CA2124593A1 - Virtual reality imaging system - Google Patents

Abstract

The virtual reality imaging system (10) takes a multidimensional space that contains real world objects and phenomena, be they static or dynamic in nature, and enables a user to define a point and/or a path through this multidimensional space. The apparatus (10) then displays the view to the user that would be seen from the point and/or path through the multidimensional space. This view is filtered through user definable characteristics that refine the real world phenomena and objects to a perspective that is of interest to the user. This filtered view presents the user with a virtual view of the reality contained within this multidimensional space, which virtual reality presents data to the user of only objects, views and phenomena that are of particular interest to the user.

Description

2 PCI`/US93/~9128 . 21124~3 qr ~aGI~ilG 2~8q~f ':.
.FIB~D QF ~E: ~IO~
Thi~ in~rention relat~ to ¢oml?uter ge~lerate~
i~aagfis ~d, iXl particular, to a ~y~tem that ar0ate~ a 5~ ual i~age of~ a~ multidi~e~ional pace to pr~en* a filt~red i~ag~ of variou~ thre~ di~e~sional phe~o~e~a a~ fQatura~ that are containe~l ~ithiD th~ :
~ult~ai~iO~al ~pac~ as ~iew~a ~rom 2~y pred3fi~6~d loou~ YithiD~ spac~.

O: :: ~ : P~OB~
~ , . . .
- :Et i~ a pro~bI~ in co~pl~:~ ¢om~ut~r co~trolll3d ~3y5t~19 t~at d-~l ~ith real ~orld ph~ao~e~ to pr~e~t repre~entatio~: of the phe~oDIe~ ~snDer that i~
: botla i~orma~i~e to the user ~nd i~l a ~i3aple 15pr~se~tation~ Sormat. Comput~r geIIQrated grap~ r~
iquitou~:: a~d ar~ t~pieally u3~dl to praC~nt ~D
a~ rate re~prQsentation of ~ object, ~ phe~om2~a, ~ultidim~ion~l ~p~a~ a~d iBt~rac:tio~ ther~bet~
Co~laputer ~e~er~ed g~rap~ic~ ~re al~o u~g~d ~:tea~si-~ly 2 0ln ~i3wl~tio~: ~y~tems to pr~se~t ~ i:laage o~ 21 r~
world $ituatio~ or a hypothetic:al ~itu~t~o~ to a u~er f~r tr~i~i~g~ a~alysis or other pu~po~ . CQmputer ~ge~erate~ graphics ha~t~ b~¢ome e:~trQmely ~ophi~ticated a~ repre~e~t ~tr~m~ly co~ple~ ~d fa~ciful 25~3ituatio~s i~ a ~a~r that i~ ~ rtually li~liXe.

~.,!,'~3~3 ~

~he application of computer graphic~ 5pan~ ~any teeh~ologiQs and applicatisns.
o~e area i~ whi~h computer graphic~ has yet to ma~e a signifi~a~t impact i8 the nrea of real time di~play of comple~ real worla phenome~a. Ro~e elem~ntary work ha~ ta~en place in thi~ area but 3ystem~ of great fl~ibility ~nd adaptability that c~n handle e~tre~ely c~ple~ pheno~ena ~r~ pre~ently una~ailable. It i~ therefore 8 proble~ to ~i~ually ~o display a ~omplex mult~dimen~ional and real time : phe~ome~a in a 1arge multiai~en~ional space i~ a ample m~nner that maps the reality to a predefined user~ viewpoi~t.

:' :~ :

:

; .

: :
:::

WO g4/0~312 PCr/US93/0~128 -`; 21245~

~OI.IITIO~
The ~o~e d~csibe~ proble~s are ~olved ~d te¢h~iz~l ~ ce nohievea i~ the l~iel~ by the ~rirtu~l reality image ge~eratioll sy~tem of the pre~ent 5 i:~ve:~tio~. Thi~ appa~atus ta~ multi~i~aensio:~al ~pace that contain~ real world object~ ph~ome~a, be they ~t~tic or dy~ in natur~, ~d e~ble~
user to d,efine a poi~t ~d/or a p~th through th~
~ulti~ime~iona1 ~pae~e. The app~r~tu~ the~ plays 10 the ~view to t~e u er that woul~ be seen ~erom the poi~t a~or p~th through the multidimen~ional spaceO Thi:~
~i~W i8 iltered through u~er aefill~ble : ~ char~ct~ri~ti hat refi~e the r~l ~orld phe~omena ~d objec:t~ to a p~r~pective th~t i~ of intere~t to 15 t~e u~er. ThiY filter~d ~iew pre~t3 the u~r ~rith ~ irtu~l ~ie~ of the r~ality c:o~tai~es~ ~r th~ this multi~liDtanqio~l ~pace, which ~irtual r~lity~ pre~e~ts a~ta :to t~l~;u~er of o~ly objact~, view8 a~d phenomena t~at ~re of p~rt~ r iutere~t to the user. Thi~
20: ~ apparatu~ highliqht~, emphasiz~, ael~te~
: reori~ts~ t~e ~ rQality contai~a ~ithi~ the :
~ultidime~qioaal ~pace to pre~e~t a~ iDlage to t~ae las~r : : ; of only wh~t ~ the user ~eed~ to ~ee to at:compli~h a ~tat~d t~qk. ~ The ~elactive pre~entatio~ of i~forD~atio~ iD ~ r~al t~e of r9al ~rorld pheno~2~a h~ u~Qr to proc eYQ t~l8 reauc~ ~ata ~et co~t~in~d ~i~ tb~ i~age pre~e2~to~ ~y thi~ ~ppara~u~ to perfor~ ~ ~e~ignate~ taslc in a ma~xler that ~e~retofor~
~r~s impo~ible.
~ The preferred ~odi~ns~lt d~oribed ~erein i~ t~at of a~ ~irport operatiG~ y3t~ vherei~ ~ airport i~
loaat~d : in a preastermiD,ed loca~ion n ~ ~:
~ultidi~eD, ionsl ~pac:e and i~ surrou~ded by variouQ
thr~ im-nsion 1 topological !lurface featur-3. ~h-' ~, -.

W094/08312 PC~/USg3/09128 212 4~9 3 three dimen~ional air ~pace ~urrounding the airport is typically managed by air tr~ffic controllers to route aircraft in the vicinity of the ~irport into arri~al _nd dep_rture pattern~ that _void the topologi¢al feature~, various weather conditio~ around the airport, ana other~aircraft that share the airspace ~ith ~ p~rticul~r flight Thi~ problem is extremely ¢omplex in nature in that the multidi~en~ion~l ~paoe around the~airport contains fiYe~ objects such as the airport ~d it~ surrounaing topologic~l feature~ a8 well ~8 dyn mic phenomena such ~g meteorological event~ that ar- beyond the control of the air tr_ffic controllers ~ell as dyn~mic phenomena, such a~ the ~ircraft, th~t c~n be indirectly controlled by the air lS tra;ffic controll-r~ The dyn~mic phenomen~ ~ary in time ana space ~D~ th~e mo~eoent of the aircr_ft within this multiai~ nsional spaae mu~t be ~_nagea in real time in response to re~l time an~ sometime~ ~udden ch ng-s i~th-~et-orologic~l phenomena _8 ~ell _8 the 20 ~ pos~tion of other aircraft No ~now~ system even remotely appro~ches proviaing tb-~air tra~ffic controller , the pilot~ or other potenti-l;users~ ~ith re~sonable distillation ; of all of the~dat~ co~taine~ with the multidime~io~al ZS ~pac- aroun~;an airport Exi~ting a~rport operation~
i~clu~e a signific~nt ~mount of aata acguisition instrumentation to provi~e the air tr~ffic controller~
~ ~ . : : .
a~ ~ell as ~th-~pilot~ of the ~ircraft ~ith ~ata relating to ~e~ther, air traffic an~ ~patial ~-rel~tion~hips~ of the aircraft ~ith re~pect to the ~irport an~ th- groun~ le~el The proble~ with this apparatus i~ th~t all of the a~ta acqui~ition in~trumentation is configured i~to indi~idual un~ts, each adapte~ to present one set of narrowly define~

:

,:~

WO 94~0B312 PCI/US93/09128 -`` 212~59~

relev~t info~atioD, to the user ~ith little attempt to integrate the plurality of ~y~tems i~to a unive~
in~tru~ent ~h~t ~n be n~aptsd t~ controll~bly pro~i~e ~n image of the multidimen3ional sp2~ce to the variou~
5 u~r~ ~ with each image being pre~qntes I to u~er in texm of their ~peci~ic nse~ for information. T~
especi~lly i~port~nt ~i~ce t~e ~ir tra~fic controller ha~ ~ sig~ific~tly dif~erent ~eed for information th~ the pilot of the ~ircraft.
lo ~he apparstu~ o~ the pre~ent i~e~tio~ obtai~
~ata from a multitude of dat~ acgui~itio~ source~ and controllably m-ld~ this i~formation i~to ~ ~ataba~e that repre~e~t~ ~11 the inform~tion of inter2st relating to this multidime~io~al ~pao~. Graphic proce-~ing apparatus re~ponds to u~er i~put to define a preaet~rmin~d point or path through the multi~i~en~io~l space a~ well a~ ¢ertain ~: visualizatio~ oharacteri~tic~ for each indi~i~u~l user. Th~ graph~c proce~i~g apparatu~ the~e, i~
real time, pr~o~t~ the u~r with a cu~tomized v~ of :~ . the multi~imen~io~al ~pace in a Yisual for~ by ~el~ting i~formation that is e~traneou~ or co~fu3ing and presenting o~ly the data that i~ of ~ig~ifica~t rel~vance to the particular u~er a~ ae~i~ed by the ~ 25 filt-r. I~ a~ ai~ort op~ration e~iroDme~t, low : ~: la~el wi~d ~hear al~rt systQ~ Lwa8) u~e grou~d-: ba~ e~or3~to gesQrate data in~i~ati~e o the :~ pre~nc~ ana locu~ of ~eteorological phenome~a 8UGh a~
~ind shear a~a gu~t fronts in th~ inity of the airport. I~ additio~, terminal doppler weather rad~r tTDWR) ~ay al~ be pre~e~t at the ~irport to ide~tify the pre~ence and locu~ of met~orological pheno~e~a i~
the region ~urro~di~g th3 ~irport to e~able t~e ~ir traffi~ controll~r~ ts guide the ~ir~rsft arou~d W094/08312 PCT/USg3/Ogl28 2 1 ~ ~ ~ 9 ~

undesirable meteorologic~l phenomena such as thunderstorms Additional data i9 av~ilable in the form of LaND8AT d~t~ indic~tive of topolosical surf~ce features surrounaing the airport Air traffic control radar is ~lso available to indicate the presence and lo¢us of ~ircraft within the ~pace around the airport for ~ir traffic control purposes Colle¢tively, the~e ~y~tems pro~ide~data represent~ti~e of the immut~ble characteri~tics of;~the ~ultidimensional sp~ce ~s ~ell ~s the dyn~mic ph-nomena cont~ined in the ~ir spa¢o, incluaing meteorologi¢~l e~ents ~nd ~irGraft oper~tions It i-~not uncommon for ~irport oper~tions to t~e~place~in~a~zero visibility mode ~herein the ;
pilot's ~bility to obt~in a ~isual image of air sp~¢e lS in front o f ~the~ircraft i9 impaired to the point ~her- the pilot i~ flying blind~ The pilot must rely on the air tr~ffio oontrollers a~d raaar cont~ined within th-~-ircraft to ensure th~t the pilot ~oe~ not fly the~ ~ir¢r~ft~on ~ collision ¢our~e ~ith a ~ol~
ob~sct, such~s~nother ~ircr~ft or the topologic~l features surroun~ing;the ~irport The virtu~1~reality imaging ~ystem of the pre~ent b ~ invention converts the ~ta obtaine~ from the multitu~e~ o~f~ ~y~t~m- ~d ~i~tills the infonm~t~on ~ cont~ined therei~ into ~ u~liz~tion of the flight p~th ~pre~ent1y~in ~front of~the ~ircr~ft Thi~
appar~tus c~ delete~e~traneou~ nformation, such a~
¢1Oud~, og,~-tc ~nd~i11u~trate~to th- pilot ~nd/or the a~r tr~ffic controller only~phenomena that ~ould :
be of ~ignific~ntlintere~t to the pilot, ~uch a8 angerous meteoro10gica1 ph-nomena~n~ other aircraft, to present the~ pilot with ~ clear image of hazar~s within the multidimensional space to permit the pilot to chart course throuqh the-- h~z~rd~ Yithout the W094/08312 PCT/USg3/0~128 21215~3 pilot being ~ble to ~ee the~e danger~ ~ith the ~ake~
eye.
The specific ex~mple note~ ~bov~ imply one of m~ny ~pplic~tion~ of this concept uhich oper~tes to S filter v~t ~mounts of d~t~ typic~lly found in ~isual im~ging situation tQ present ~ "clearer image"
to the user ~ ~define~ by the specific needs of the user. ~he user therefore sees only ~h~t they need to see ~na c~n ~complete t~sks th~t heretofore ~ere ~ impossible due to the visu~l overlo~d encount~red in ~y ~itu~tions, such ~8 flying ~n ~ircr~ft through fog or cIouds or~not being ~ble to i~entify ~ ~ind she~r event~ in~ a meteorological phenomena of signific~nt ~tent ~nd c~mplexity. An a~ition~l S c~p~bility of this y~tem is the prediction Qf future st~tes of the~dyn~ic pheno~en~. D~t~ is collected by the~ ~ultitud-~of~data ~cguisition system~ over plur~1ity ~Or ~ampling interval3 ~d c~n be apolate~ to~illustrate the state of the dy~amic 20;~ ~ ph-noo-n~ ~ ; f~ture sampl~ng int~rval~. Th~s cap~bility~nabl-s the air traffic co~trol supervi~or to~modQl the~eather activity around the airport to pro~id-~infor ation to plan airport operations for the immeaiate~future. -~

;: :~: : : : :: .:

:: : : ~ .,;, .

: :

21~4~93 BRI~F D~8CRIPTIO~ OF TEE DRA~I~G
Figure ~1 illustrates in block di~gr~m form the overall architecture of tho app~ratu~ of the pre~ent in~ention:
Figures 2~ lllustr~te in flow di~gr~m form the operation of th- ~arious ~egments of the improved ~eather alert system;
Figure 5 illu9trate~ in block aiagr~m form the overall srchitecture of the improved ~eather alert 10 system;
Figure 6 illu~trates a plot of a typical ~irport conf~iguration,~incIu~ing ~L~A8 ana TDWR in~t~llation~
~nd~typ~cal~è~th-r cond~tion~;
Figure~ 7 ~ 12 illu~tr~te ~n e~ample of lS con~erting~ ~the compact dat~ repre~e~tation of ph-nom-n~ to~ ~;a three-dime~sional object represent~tion~ d Figur~13~ 17 illu~tr~te typical vi~ual image~
pro~uce~ by ~t~ appar~tu~

: ` ` : :

.

WO g4/08312 2 1 2 ~ 5 9 3 Pcr/US93/09128 _9_ DBTAI~D DE8C~U:~IO~
Figlsre 1 illustrates i~ bloclc diagra~ fom~ the overall xrchitQeture of ~h~ virtual reælity im~gi~g ~yst~m lO of the pre~ent inve2~tion. lErithi~ the S virtual reality i~gi~g sy~te~ lO, a dat~ acquisitio~
ys~tem 1 fu~ctio~s to coll~ct s~d pro~ce th~ real t~e ~at~ thst is repre~e~tative of the multi~ime~ional ~pac~ ~d the feature3 an~ ph~ome~a extant therei:l~. Graphic~ 3ysteDl 2 i~u~ation~ to utiliz~ the real time data that i 5 produ~e~ by the datz~ acquisition ~ y~t~ 1 to produce the ~ ual ~isplay~ required ~y the plural~ ty of user~ . To accompli~h this, a bared datab~ 3 is used i~to ~rhich the real tiDIe data i~ written by the data ~5 acqui ition ~ yQtem 1 and acce~ea by the variouQ
proc~ing ~e~ent~ o~ g~aphic~ ~ubsyst6~ 2., A u~r :dat~ input device 4 i~ pro~rided to e~able a u~r or a plurality of u~er~ to ~nter ~ata i~to the graphic~
sub~yst~ 2 indicst e of the particul~r illformation : ~ 2 0 that: each of t31e plurality of usar~ ~e ires to ha~.re ~lisplayed o~ the o orre~po~ding aisplay de~ic~
operatio~r the data acqui3itio~ 3ub~3y8tem 1 ¢o~prise~
a plurality of ~data acguisition apparatuQ 21-2n, eac:h of whic:h produae data rep esentative of mea3ur~ent~
,: ~ . .
25 ~ perfor~d ~o~ ~ t~e phe~om~na or featur~ th~t are located in the~ nlultiai~e~io~al 8paOQ. ~rhe~e dalta acgui~ition ~pparaLtu~ 2sl a~ pro¢~ the real tim~
measureme~t ~ata into comp~¢t data r~prese:~lt~tio~ of th~ ph~no~e~a s~d f~ature~, u~ ompa~:t data 30 xepr~ssntatis~ ~re tra~itted to graphic~ ystem 2 or proce3sing i~to th~ Yi~ual i~gss. The graphics 311b y~te~ 2 :~o~ert~ th~ compac:t data repre~tations produced by thQ plurality of data acquis~ition ~Ipp r~tu~ 21-2n into visu~liz tions ~ d~f:L~0d by e~ch W094/08312 PCT/US93/091~8 9 3 -lo-of the u~era of the Yirtual reality imagi~g sy~tem 100~ ~hi~vi~ualization i produced ~y performing databa~Q tr~sver~al to pre~ent the data i~ a for~ ~nd format of interest to e~ch of the uaers.

Aviation Weather DisDl~Y 8Y~tem A typic~l ~pplioation of thia app~ratu~ ia ~n ; ~i~tion w-~th~r~diapl~y syatem whoae d~ta a¢qui~ition ubaystem~ m~k- ua- of ~ plurality of aviation weather in~tru~entation that are u~ed i~ ~nd ~bout ~n airport lo inat~ll~tion.~ The~ ~viation we~ther i~strumentation ; m~y in¢lud- ground b~aed ~enaors such ~5 rad~r, lighting ~d-t~ction~networ~, an~ wind ~enaora ~ well a~ airborne~s-n~ors ~uch ~8 aounding balloons or ~; aircraft ba~ed~en~ors E~ch of the ~iation ~eather 5 ~ in~trumentation~produc-3 r~w data i~dicative of re~l time m-teorologi¢al phenomen~, topological fe~ture~
a~ aircr~ft~op-rations in the multi~i~en~ion~l sp~ce, ~hi¢h~real ~time data is pro¢e~se~ by the ~ata acquisition~ s~ system 1 to produo~ compact 20~ ~e~rosentations~of the re~l time ~t~ Theso ~ata proc-ssing~ 8tep 9 ot-n include filteri~g, feature ~tr ction~ ~nd~correlation/int-gration of more th~
' one~at~ 8tre~m~ Furthe re, thi proce~ed data may ' be~';use~ input to physically based model~, ~hich 25~ att ~ pt~to~predict~the evolving ph-nomena ba~ed o~ the 'stor~e~ureme~t~ ~
Fro~ th-~compa~t~ dat~ represent~tions, the graphics sub~y~tem~2 generate~ ge~eralize~ graphio~l repr-sent~tion~of th- phenomena ~n~ feature~ Thi~
inqol~e~ the creation of ~n object or object~ which e~i t ~n a ~irtual multidim-n~ion~1 p~ce In an ~viation ~eather di~play application, this virtual reality imaging ~ystem 10 must op-rate in real ti~e :::

W O 94tO8312 2 1 2 4 5 9 3 PC~r/US93/09128 since 3ignific~tly del~yed data ~ffects the validity and functionality of the ~y~tem a~ ~ ~hole The visualization pre~ente~ to the user typically ~nclu~es frame of referenoe information such a9 terrain, S overlaia ~ith iaentifi~ble features in the form of highways, r~nge rinqs or icon~ representing ~nicipalities or airports Furthormore, the terra~n surface c~n be~colorea ~y te~ture mapping it ~ith ~n .
image such a~ ~LAND8AT image or a digital map In or~er to integrate the plurality of data streams that ; are produced in a~data acqui~ition subsy~tem 1, the graphics sub~ystem 2 must perfoxm numerou~ operations uch as database~culling, relative level of detail detexmination ana rendering to oreate user lS reoognizable images from the xaw data or compact data repr-sentat ons~that are stored in database 3 `~

ata Acauisition~8ubsystem Archi~tecture ~ Figur 1~illustratQs the m~jor ~ubcomponents of ; ~ typical data~cgui~it~on apparatu~ 21 In ~ typic~l ~20 configuration,~a plur~l~ty of sen~ors 201 are use~ to ma~e ~easur _ents during a sampling interval of pr-d-t-rmined;~duration ~nd repetition freguency, of one~or~mor-~char~cteristics of a p~rticular phe~omena ; or~f--tur- within the multidi~ensional ~pace The 25 ~ output signals from the plurality of sen~ors 201 are r-ceived by~ data ~filtering ~nd feature extraction elem-nt 202 ~hich functions to filter the data re¢-ived from the~plurality of ~ensors 20 to remove ~mbient noi~e or unwante~ signal co~ponent~ therefrom ~ Th- data filt~ring, feature o~traction element 202 ~al90 function~;to convert the raw ~ata r~ceived from the plura~ity of sen~or~ 201 into a definition of the parti¢ular ~phenomena or feature that is being .

WO94/0~312 ~ l2 ~ PCT/USg3/09128 ~onitored ~y thi~ pzrticular d~t~ acqui~itio~
~pparatu~ 21. A~ e~amplQ of ~u¢h a capability is the u8e of ~n i~proved lo~ }eYel ~ind ~h~r deteGtio~
apparatus which convert~ the ~i~d ~agnitude me~ur~ment~ from a plur~lity of ground b~e se~or~
into dat~ repre~e~tative of ~ind shear eve~t~ ~ithi~
the multidime~sional ~p~ o ac¢o~pli~h thi~, the r~w data obtained fro~ the sen~ors 201 ~ust ~e co~verted i~to a form to e~tract the ~ind ~he~r events from the plurality of ~ind ~e~surement~ tzken throughout tha ~ultidime~qional space. Ths re_ultant informatio~ i~ used by compaet data repre~e~tatio~
: apparatus 204 t~o produce a SQt o$ ~ata i~dicative of the ~xtracte~ fe~ture i~ ~ con~e~i~t ~ory ef~icient m~nner. Thi~ c~n b~ i~ the form of gridde~ d~t~ ~et3, :: : $o~ture ~t~nt a~a location ~at~ a~ well a-~ other : po~sib}e repr~e~ation~0 Furthermore, the data ~oguisition ~pp~ratus can i~clude ~ predictive elQ~ent 203 which UQ~ th~e d~ta obtai~ed ~rom ~ata ~ilt~r~g, .-, :~ 20 featuro e~tr~ctio~ apparatu~ 202 to e~trapol~te i~to : : one or ~orQ pre~te~ined ~uture ampling i~ter~al~ to identify ~a $utur~ temporal ~tate Or th~ fe~ture or phe~om~na that:i~ b~ing mea~urea~ The dat output ~y the predicti~e~ ele~e~t 203 is al~o for~ar~e~ to comp~ct ~at~ repr-~e~tation ~le~e~t 204 ~or i~lu~ion in the d~ta J~t t~at i8 produ~e~ there~. The r~ ult~ut co~pact aat~ represe~tat~o~ are tr~mitte~
to the grap~ s ~ y~t~m 2.
It i3 ob~iou~ that if th~ f~ature b~i~g mo~itore~
i~ tempor~lly ~d spati~lly st~tic, th~ ~ta that i~
produce~ i8 inv~ia~t a~ ~ee~ not be upaated ~uri~g succes~ive ~mpling i~terval~. ~owever, m~t phe~ome~a that are mo~itored i~ this environme~t te~d to bs temporally a~d in many ca~e~ ~patially ~aryi~g WOg4/0831t PCT/US93/09128 212 4 ~ 9 3 ~n~ the operation of the ~ata acguisition apparatus 1 i8 on a ti~e sample~ basi~, with a set o~ ~ata being produced at the end of each sampling int~rval. The plurality of ~ata ~cguisition elements 21-2n preferably operate in a time coordinate~ manner to produce synchronized sets of data set-~ in the database 3 80 that graphics subsy~tem 2 can produce temporally coordinatea ~vi~s of the phQnomena and features located în the multiaimensional space on a once per sAmpling intervàl~ basis or over a plurality of ,~
sampling int-rvals, dependent on the amount of data that~mu~t be prooess;ed. In a real time e~vironment, the~ plurality~of~data acguisition apparatus 21-2n ~functio; to collè¢t tremendou~ ~mou~ts of data ~nd ~;~ 15 reduoe the~data~to manage~ble ~mounts for use by the -~
graphics subsystQ~ 2. ~ ;
The improve~low-lev-l wind shear alert system, u~trated~ in~block diagra~ form in Figure 5, pro~iàés an`~improved method of identifying ~he rQs~enoe an~locus of wina ~hear in~ a pr~dQ~ined area.
This low-level d nd shear alert syst~ enhance~ the operational effectivenes~ of the e~i~ting ~WA8 sy~tem by~appi~g~th~two-dimQnsional wind velocity, m~asured ~t a nu~ber of looations, to ~ geographi¢al indi¢ation 2~5~ of~ind she~r events. This re-ultant geogr~phi¢al indication is;di~spl~yed in~ color-graphic form to the ~ir -~tr-ffi~ ¢ontrol ;p-rsonnel ana can also be tr~nsmitt~d~vi~ a~telemetry~l~ink to ~ircr~ft in the vioinity o~f~the ~irport for ~iQplay therein. In aadition, gust~fronts are trac~ed ~a their progress through th-~predsrined~-r-- displ~yed to the users.
This low-le~ol~wind sh-~r alert system c~n ~180 ; i~tegr~te~d~t~nd processea inform~tion recei~ed from a plurality of sources;, such ~s ~nemometer3 ~na :

::

WO94~08312 2 ~ ~ 9 3 PCT/US93/U~1~8 Doppler radar systems, to prod~ce low-level ~ind 3hear ~lerts of signifi¢antly improved accuracy over those of prior syste~s In particular, the apparatus of the i~proYed low-level wind shear alert system makes use of the data ~nd~processed information produce~ by the existing Low-~e~el ~ind ~hear Alert ~ystem ~LLWA8) ~
~ell as that produaed by the Termin~l Doppler We~ther ::
R~dar ~TD~R)~ to~ pr-ci-ely identify the loeuq an~
magnitu~e~of~ lo~-le~el ~ind shear event~ ~thin a predetermi~ed are~ This is accomplishe~ by the use of a novel-~i~tegr~tion system that utilizes the ~at~
~nd~ proeessé~ information recei~ed from these two -y-tems~ ~LLW~8~ TDWR) i~ such a way that the limitatio~s~ of the two ~t~nd-alone systems are 5~ ~meliorato~ This integration soheme, ~hile a~r---ing~th--e~li~it~tions, simult~neously maintains th-~trength-~of~th-;t~o stan~-alo~e ~y~tems This techn~que~th-n~provides the best possible wi~ sh~r h~z~r~ lert~ information Further~ore, thi~
2~0~ i~tegratio~ ~ tho~ology adare~se~ the oper~tor i~ter~ction problem ~iscus~e~ ~bo~e The integr~t~on is~ ~fully~aut~o at~d, reguire no ~eteorologiGal interpretation~by the users ~ d produce~ the xeguired graphical ~d~ alphaDumeric inormation in ~n 25~ unaJbiguou~ ror~at ~ ~ L~tly, thi~ i~tegratio~
tech~;~que~ ~ lemented fully without any ~ajor software~ mo~if~ication nor ~ithout z~y har~ware modifications to~the existing stana-alone ~y~tem The TDWR~apparatus u~es a S cm C-band Doppler radar sy~t~m~to me~sur0 radial wind~ when ~tmospheric catte~rers~aré~ present This sy~tem processss the radar return~ ignals to create a field of radially oriented line~ ~eqments indicative of the radial velocity data~ received from the radar The TDWR

:: : : ~: :~;

: :

:

WO 94/08312 2 1 2 ~ 5 9 :~ PCl/~JS~3/09128 appar~tus bound~ i~olated s~ts of ~egme~t~ that ~re ~bov~ a pr~etermin~a thre~hold to ~fi~e an srea which ~ould contain a specifi~:, potential low-level ~ri~a ~hear e~re~lt. The bounaing i~ such th~t it 5i~lcorpor~teQ th~ :~mallest area ~rhieh i~lud~s all of the line sQgm~nt~ abov~ th~ predetQrmi~e~ threQhold.
A pr~liefined g~o~etric _hape i~ UQe~ to produce l:hi~
bou~l~i~g ~d, the characteri~tics of thi~ geometric ~hape are adapt~d i~ order to ~nc:ompas~ ~ll ol~ th0 lbrequir~d aata point~ in the ~ini~al are~.
The app~ratu of the improved lo~-level ~ri~d ~hear alert ~y~tem i~ di~ided i~to t~o i~depenae~t section~: dete¢tio~ of wi~d ~hear ~ith lo~ ~ituatio2s (microburst~, etc.) ~nd detectiQn of wind 3hear with 15gai~ ~ituatio~s (gu~t fro~t~, etc.). ~he TDWR ~ystem o~tput~ ~ind 3he~r with lo~ data i~ the form of : microburst ~h~pes. The e~hanced low-leval ~i~d ~hear :alert ~y~tem gen~rate~ eguivalent hL~ ~ioxo~urst sh~pes u~ing th- trla~gle ~nd eage ~i~erge~e ~alues 0 :produced by 4he e~i~t~g ~WAB apparatu~. The ~wa~
: : ~i¢roburst ~bape~ ~re vali~ated by u~ing aux;li~ry in~oxmation from ~WA~ ~ d TD~R to Qliminatæ mar~inal and al~e-detectio~ ~W~8 mi~robur~t shape~ The ~: rs~ultant two s~t~ of mucrobur~t sh~peq are then : 25¢o~siaer~d ~or alarm ge~eratio~ purpo~e~. ~he ~iad ~he~r~with galn portloa of thi~ ~yst~m simply di~id~s the ~ower~ge~ ar~a i~to t~o x~gio~3, ~ith ~D~
produci~g wi~d~she~r ~lth gai~ run~ay ~l~rm~ for ~ind ~hear eve~t~ tb~t o¢Gur out~ of the ~wa~ ~a~or 30while the LLWa~ ru~ay ori~ted gain ~larffl~ ~r~
produc~d for wind ~hear eve~ts th?t oc~ur i~i~e of the ~L~A~ ~e~sor ~etwork.
~hi~ integr~tion architectur~ e~ble~ the co~curre~t uqe of ~ plurality o ~ensor-ba~e~ syqt~

WO94/08312 PCT/US93/Ogl28 2 1 ~ ~ ~ 9 ?

to pro~ide the wind ~hear detectio~ function ~ith incre~sed accuracy. Both ground-ba~ed ~nd air¢raft-base~ ~ensor systems c~n be u~e~ to provide ~ind dat~
for thi~ ~pparatu~. The m~pping of diverse form~ of S input ~ata into a co~mon data structure ~predefi~ed geometric shape~) avoids the neces~ity of modifying e~isti~g ~ensor ~yJtems and simpl$fiQs the pro~uction ~; of information ~i~pl~ys for the u~er. ~he use o~ a com~on i~form~tion di~pl~y apparatus ~n~ fo~mat 10~ ren~ers th- combin~tion of sy~t~ms tr~n~p~rent to tho user. ~ ~
: :. , ~ , I prove~ Lo~-Level ~iDd 8he~r Detection 8Y~teQ
Adverse~ ~eather conditions, e pecially those ~ .
affecting airport~operation, are a signific~nt ~afety 15~ conc-r~ for~irline op-rators. Low lev~l ~ind shear is~ of ~ignificant intere~t bec~u~e it ha~ cau~ed a ;nu ber~of m-jor~ir ¢arri-r accident~. ~in~ ~he~r i~
a~ch~g~ in~in~ spee~ ~n~/or ~irection between ~
two points in~the~at~osp~ere. It i~ generally not a :20~ :: erious~hazar~ for aircraft en route between airports at~:normal cruising altitudes but strong, sudden low-le~el wi~d ~s~ear~ in~the terminal area can be ~ea~ly ;for~n~ircr~ft~ o~ approach~or departure from ~n ai:rport.:~ The~:~most:~haz rdou~ form of ~ind ~hear is the :2~5~ micro~urst;, ~n~outflow of air from ~ sm~ll scale but powerful downward gush of col~, heavy air that c~n occur'~benéath~or fro~ the storm or,rai~ shower or even i~ rain freè~:~ir un~er a harmless looki~g cumulus ; cloud. As thi~s:dow~araft reaches the earth'~ surfac~, ~its spreads;~out~ horizontally li~e a ~tream of wator ~ spxayed ,trai;qht down on a concrete driveway from a : ~ ~: : : garden hose.~ An aircraft that flie~ through a : microburst~at;~low:altitudQ first e~counters a ~tro~g ~ ~ :

:, .

212~ 3 hea~win~, then ~ ~owndraft, an~ finally a tailwind that pro~u¢-s;a sharp re~uction i~ air spee~ an~
~u~en 108~ of lift This loss of lift can cau~e ~n ~irplane to sta11 ~a~ cra~h ~hen flying at a low S spee~, ~uch as ~hen ~pproaching ~n airport runway for l~n~ing or ~departi~g on ta~eoff It i~ therefore ~ desir~ble to prov~de pilots with a runw~y ~pecific ; ~ alert when a fifteen ~not or greater hea~win~ 10~8 or ga~n situ-tion;~ etectea in the r~gion ~here the aircraft are~below one~ thou~and feet ~bo~e grou~d level ~na~ within three nautical ~iles of the runway Figure 6-~i11ustrate~ a top view of a typical airport installation~wherein the airport i~ within the region indi¢?tea on~the horizontal a~is by the line labelod~ana~a~T-rminal Doppler Weathor Rad~r ~y~tem ;502 is locatèd~a~;di~tance D from the poriphery of the ai~port Inciua a~ithin~the boun~s of the ai~port are plurality~ Or~ ~ Low Lev~l ~ind 8hear Alert 8y~te~
2~0~ ensors 505 ~ The sensor~ 505 are typi¢ally an-Io~t rs;;loc~ted;two to four kilometers apart ~nd are~u~ed to~pro~uce~a single plane, two dimen~ion~l pictur- of~the wind~v-locity within the region of the ~irport ;Th~-Term~n~l Doppler ~eather ~d~r 502, in 25~ contra~t, consists of ~ one~dimen~ion~l ~r~dial) bea~
which~Jc~ns~ runy~ys~ R~ and flight p~ths but can~aeasur-~only ~;radi~l horixon~1 outflow component of~-win~ The;~omi~ TDWR~-o~n ~tr~t-gy produce~ o~e urface elev~tion 8c~n per mi~ute ~nd ~c~ ~loft of the ope~tio~a1~r-gio~ to an ~ltitu~e of at lQ~st tw-nty thous~D~f--t every two ~nd a half minutes Thi~ strategy~is`~inte~de~ to provide frequent up~ates of surf~ce~outflow while monitoring for fe~tures ~loft ; to i~dicat-~ th~t ~ microbur~t i~ imminent : .

: ::~ :
:: :

WO9~/08312 PCT/US93J09128 2124~9~ 18 ~icro~ur~t~ (M1-M8) are re~og~ized primarily by surf~ce outflow although they cah be ~ticipated to ¢~rtain extent by monitoring feature~ a~d ~vent~ in the region above the ~irport loc~tion.
S Thunder~torm~ typia~lly produce a powerful downward gush of cold he~y air which spre~d~ out horizontally a~ it reache~ the e~rth'~ ~urface. one ~egment o~ thi~ downflow ~pread~ out ~w~y from TDWR
rad~r ~hile ~n opposi~g ~egment ~preads out towar~
tho TDWR radar~. It i8 generally a~sumed that these : ; outflows ~r~ ~ymmetrical for the purpo~e of ~etecting ~icroburst wi~ ~hear~. Becau~e most microbur~ts do not have purely ~ym~etrical horizontal outflow~, the TD~R sy~tem can ha~e problemq detecting or estimating : 15 the true inten~ity of a~ymmetrical microbur3t outflows. A can be ~en fro~ Figure 6, the ~nemometers 505 of the Low ~evel ~iad-~hear Alert By~tem;ar- sIt-a~on both .ides of ~irport ru~w~ys Rl-R~but do;not exten~ to the full thx~e mil~ t~nca 20~ ~from ~the~ end~ of the runw~y ~ desirable.
: Therofore, the ~an~omet~r~ 505 can only ~etect ; horizontal ~irflow3 that occur in their imm~iate : ;~ioinity ~M2,~ M3, M5-N8) even though th~re can be horizoDt~l airflow outside th~ ~emometer network (Ml, : 25 M4:~:that ¢an~imp~ct airport op~xatio~s but are out~ide of the-r~nga o r the limite~ nu~bsr of anemomet~r~ 505 ~ ite~ ~t a~ airport.

: I~Prove~ hear ~lert ~v~te~ Axchitectur~
: Figure 5 illu~trate3 i~ bloc~ diagram form the : 30 o~erall architecture of the i~pro~d low-lev~l wind :
~hear al~rt qy~t~m 100. This lo~-le~el win~ ~hQar alQrt ~y~t~ 100 integratQ~ the grounfl leval ~ind d~ta collocted by on~ ~t of statiollary ground levol ~r.~or .

W094/08312 2 1 2 4 ~ 9 3 PCT/US93/09}2g ~anemometers) 505 with the higher altitu~e wind d~ta collectea by ~ ~econd se~or IDoppler radar) S0~ in order to accurately identify both the locu~ ~nd magnitu~e of low-level win~ shear condition~ within a S predetermined area A The two ~ets of dat~ input~
illustrated in this Qmbodiment of the inve~tion include the aata produced by exi~ting data proces~ing system~ ociated with the sensors in order to preprocess th-~aat~ prior to integration into the unifio~ precise output presented to the end user The sensor sy~tems include the eYisting Low Levol ~ina 8he~r Alert 8ystem ILL~A8) front en~ procossing ,: ~
101 which i~ ~an ~nemo~eter-ba~ed wi~d shear alert system u~ed to~detect the pre~ence ~n~ identify the locu~ of wind shear events at or near ground level Th- LL~a8 s~ste~ 101 generate~ data indic~tivo of the wind velocity~(m~gnitude ~nd direction) ~t each of plur~1ity ~of~fi~-d ~ite~ 505 located within pr-a-fi~e~ ~r- ~ The collected ~ina velocity d~ta is thQn~preproces~ea~by the LLWa~ ~yste~ 101 to identify tho locuJ~ana~ gnitude o~ wi~a ~hear~ at ground level by~ide~tifyi~g the divergence or converge~ce that occurs~in the~measured wind velocity throughout the predefi~ed~are~ 8imilarly, the ~econd set of ~e~sors 25~ ~is the ~erminal~D~oppl0r We~ther Radar (TDWR) 502 ~hich u~ Doppl-r radar~sy~tem to ~easure low-level ~ind she~r ~acti~ity~ in the predefined are~ The TD~R
ystem 502~searches it~ r~ar sc~n ~or 3eg~ent~ of the ra~r be~m ~ of~ monotonic~lly increasi~g raaial .
~elocity The~e ~egion~ and ~re~ of radial convergence ar-~identified a~ the locus of wi~d ~hear eve~ts ~The ~ntegration ~ystem 103 that h~s been ; de~eloped lor th- intagration o~ TD~R 502 ~nd LL~A~ ~

' -:
, ..:

W094~08312 212 4 5 9 3 PCT/USg3/09128 s -20- ~

1~1 u~e~ ~ product-le~el technique ~n~ ivide~ into two independent se¢tions: the detection of ~indshear-with-loss situations ~microburst~, etc.) an~
win~hear-~ith-g~in situation~ (gu~t fro~ts, otc.).
5The outputs from the uind~hear-~th-loss portion of the TDWR system 502 are microbur~t shapes - ~hich ar- use~ both -s qraphical information ~nd to gener~te the te~tual run~ay a~lert~. A~ ~n lntegr~tion "ad~-on"
to the exi~ti~ng~LL~A8 sy~tem 101, ~n enhance~ LLWA8 10section 102 wa8 develope~ to generate LLWA8 microburst ~hapes. Thes~shapes are computed u~ing tri~ngle ~n~
ge ~i~ergence~alu-s obt~ined from the LLW~8 system 101~. ~Even though~the method~ u~ed to gener~te these shape~ i8 quite~ifferent, these LLWA8 microbur-~t ~sbapes are~i~entical -;in both form an~ ¢ontent - to tbe~TD~R ~icroburst Yhapes. Thi~ allows for the s~me ,: ~:
~lert-generation~logic to be applied, ~nd for the oommon grap~ical~di~play 116 of ~icrobur~t detect~on~
Thè $D~LL~a8 (windshoar-~lth-loss) microbur~t 20 ~ ~ int-grat~on 114~is e~sent~ally the combined use of microbur~t shape~from each suk-~ystem 112, 502 This oombination,~;however,~i~ not a 8patial merging of the m~hapes each~sbape i9 co~idere~ a~ a~separate entity ; Further~ore,~ th~ LLWAS~microburst shape-~ have been 25~ pa8sed thr ~ ~a~vali~ation proce~s in ~ymmetry te~t 13~ 8y~this ~e~mean th~t ~uYili~ry inormation 703 from both TDW~ ~n~ LLW~8 ~ utili~e~ in ~n attempt to eliminate~certai~ of the ~weaker" L~WAB microburst ; shapes - on-s~th-t could generate nuisance or false 30~ alarm~ The motivation and implementation for this procedure is~described below ~o~eYer~ an alternative to this prooes~s, the en~or data from ea~h of the ~ub-system~ 112, ;~502 could b~ mergQd to produce a composite set of ~hape~ indicative of the mergea data :

~12~S93 W094fO8312 PCT/US93/09128 ~hi~ altern~tivQ proce~ noted harein in the -~
context of thi~ ~y~tem realization.
~nce a set of microbur~t sh~pe3 ~re produce~ by the enhance~ ~L~Ag apparatus 102 ~d integration apparatuY 103, the~e shape~ are tr~n~itte~ to the Terminal Doppler Weather Ra~ar ~y~te~ 502 which contain~ the runway lo~s alert generation pro¢e~.
: 8imilarly, the integration ~pparatu~ 103 receive~
: ~WA8 runway or~ented gain aata and ~W~ gu~t froht ; 10 d~ta in gu~t front integration apparatus 115. ~he : ~LWAS runway-ori~nted-gain data includes data front ; trac~ing ~y~tem 1~19 which u~e~ the L~WA8 z~em~eter ::~ : wind veotors to detect, track, ~nd gr~phioally di~play gust~front within the predetermi~ed ~rea. ~W~8 : 15ru~way-ori~nted-g~in ~ROG) i~ also u~ed for detection of generi~ wind~haar with gai~ h~zard~ ~ithin th~
LLRP8 network.: ~ ~hi3 i~ ~ot nece~arily tied to a :specif~c:gu~t front detection. Win~ ~he~r with gain : situat~on~ ca~ occur independently of gust fro~t~ -e~.~g.~the l~di~g~ -dge of a ~ic~obur~t outflo~, or rger-~¢ale ~meteorological~ ~ro~tal p~age. T~
sel~atet d~t~ is~:~t~en tr~nsmitted to the TDWR ~y~tem 505 ~h~re a:~ru~ay gai~ ~lert ge~er~tion pro~es~
: : produce~ nn ~l~rm i~dicati~e o~ the presence of a winfl ::~ ~hear~ith q~in~h~zard.
Alarm ~rbitration proce~ ~n TDWR ~y~tem 50z elect~ the:~l~r~ produce~ by either runw~y 10~8 ~lert generation proces~ or runway gain alert gen~ration : pro~es~ to pre3~nt to TDWR di~play~ 116. Th~ existing ::: :
: ~ 30 ~i~plays 116 ~:o~sist of the TD~qR Geographic ~itu~tion ~: ~Display ~GSD)~ ~rhich illustrate i~ ~raphical form the ` ~ microburst ~hapes, gu t front~ ~a indicate~ whioh ~`
runway~ ~re in ~ alert statu~. The ~DWR an~ ~A~
Ribbo~ Displz~y Tor~ RDT~ giv~ ~lph~ tric ;
'`
, 21~ g3 message ind~c~ting ~lert statu~, event type, location an~ magnitu~e for e~ch operational ru~w~y.
It i~ ob~ious from the ~bove ~escription that the e~isting LLWA8~ 101 an~ TDWR s02 ~ystem~ are utilize~
~ much ~ po sible ~ithout mo~ification to ~ini~ize cost an~ impact oD eYi~ting in~tallation~. It i~ al80 possible to impl~ent those features in other ~y~tem configurations. Any other data ¢olle¢tion ~ystQm c~n be si~ll~rly i~t-gratQ~ ~it~ the Q~isting TDWR ~ystem 502 or the Q~isting~LWAg sy~tQm by the application of the ph~losophy ~de cribe~ ~bovQ. For e~ample, the a~tion of ~nother Doppler ra~ar, or ~nother ~nemometer networ~.

8haDe Gener~tion Philo~oDhy 15 ~ ; The~L~A8~mi¢roburst sh~pe c~mput~tions are basQ~
upon the ~-tection Or di~ergence in the surface ~in~s.
These~tri~ngle~ ~na edge ~ivergence e8timates ~re Appe~ ~onto a~r-ct~ngul~r gr~ Contiguou~ "clump~"
of ~bove-thr-~hol~ grid points àre oolle¢ted ~nd then 20~ us-~ to~generat~ iorobur~t shapes Compe~sating for the ~pati~l~under-s~p1ing of the true ~urfac~ ~ind f1~ inh-rent~ i2 the~ L~WAB; d~ta a "symmetry hypothési~ u~ed~ in generating the location e~t-nt ~ and~ gn$tude ~108~ ~ ~e~ti~ate) for the~e 25~ micro~ur~t~bap-s ~ Thi~ h~ypoth~sis is applie~ a~ if a~-y~-tric~microburst ~ere cent~red at Qach (a~ove ;tbrQs~old)~ grid~ point~ In~ g-ner~l mic~obur~t ` outflow~;ar ~not ~ymm-tric ~o~e~er the spatial superposition of the~e ~ymmetric '~grid-point-~icrobursts~'~ n~a given clump doe~a very good job of appro~imating a non-~y~ tric event While a given detected di~ergence may be real the LLW~S data~ alone cannot be u~ed to determine :: :: : ` :

;

: ~
: ~ :

, . .

~hether ît is truly a~sociatea with a microburst.
Therefore, the ~pplication of the 3ymmetry hypothesi~
~ay not always be ~alid. The problem i~ two-side~.
~f the ~ymmetry hypothQsis i~ alway~ use~, it coula generate false al~rm~ in certain ~on-microburst ~ituation~. For Yample, strong ~urface winas setti~g up in a persi~te~t aiVergent pattern. on the other h~n~, if the 8ym~etry ~ssumptions are never u~ea, wind shear warning~ for ~alia microburst eve~ts could be del~yed, in~ocur-te, or even eliminated. The issue is ;~ then to deter~ine whether a given LLWA8-detected divergence i3~-ssoci-tod ~ith a ~icrobur~t ~nd henoe determine whether the symmetry hypothesi~ should be ~pplied.
15The ~lgorithm th~t ~a~ dev~loped combined "featuros-aloftN~;i~formation from TDWR: three-di~en~io~al refle¢tivity ~tructures ~nd micro~urst precur30r3~ oth proje¢ted down to the ~urface): ~nd detected ~stro~g" ~urface divergence ~microbur~t 2~0 ~ sh~pes); from~bot~; TD~ 502 ~nd ~WA8 101. Thi~
information i3;~th-~ ~ynthesized, both spatially ~d tempor-lIy to;~cr--te ~ ~et of geo~etric dis¢s. The intent~of the~ aisc3 i8 to indicate a region of the atmo3pher~ ~ithin and/or ~bove the disc, ~i.e. a 2~5~ cyl~n~er~ ~here ~ there ~ good li~lihood of ~icroburst~activity. ~This "r-gion" coul~ be in space:
th- ~tection~of~t~e surraoe outflow, or ~i¢roburst featur-~ ~bov8~ the 3urfac~ ~reflectivity ~/or locity sigDatures). It coul~ ~lso be in time, t~at is, a microbur3t is either: going to occur~
progress, or~ha3 recently been pre~ent.
Th~se;~i~c3~re then e~amine~ for "clo~ne~s~ to ~- ~those LL~A~ ~icrobur~t ~hape~ that are to be ~alidated. If thi~ pro~imity criteria i~ met, the , W094/08312 PCT/US93/09~28 2124~3 -2~

LLWa~ microbur~t shape i8 "~ali~ated" ~nd p~ssed onwar~s. That is, the u~ of the ~ymmetry hypothesi~
is a~s~ ed to be appropri~te in thi~ case, ~d this L~wa8 microburst sh~pe i~ to be u~e~ for gener~ting ~ina she~r ~arnings and to be displ~ye~ on the G8D.
If the pro~imity test fail~, the ~LWA8 sh~pe i~
~iscar~o~. ~~e~er, in thi~ latter cir¢umstanc~
there coul~ be a valid ~in~ shear h~zar~ occurring that is not a~sociate~ ~ith a microburst - or po~sibly a ~i~croburst~ehat i- not being correctly identifie~ in t~e symmetry disc c~lculations. To prevent thi~ type of ~issed~etection, the LLWa8 Runway-Oriented-Loss (ROL) information 703 is then usea ~8 ~ fall-back to generate~any~appropriate ~ind ~hear warnings.

::
~ ~nhance~LL~o~8yste~-preDroce~ d ~q Th- nh~nco~LL~A8 system creates a gri~ point t~blo for;use in creating ~icroburst shape~ This proc~-~s~ illu~trated in Figure 3 ~n~ is acti~ated At y~tem initlaliz~tion As ~ prepro¢essing step, a set 20~ of pointers~are~g-n-rated ~hich map triangle a~a eage ~icroburst~d-t-ction areas to ~n ~alysis gri~
During real-ti~e~;operation, LL~A~ triangle ~nd odge di~rge~ce~values ~re then ~app~d onto the grid -pplying a m~g~itu~e value at each grid point This 25~ -t~of grid point magnitudes ~re used with the clumps produ¢-d~by~¢lump ~h~pe gener~ation apparatus 111 to ;go~erate~ t~of low level ~i~a ~hoar alert sy3tem microburst~shapes The "pointer~" for tho mappi~g of trianglo and -dg-s to the grid i~ a "first-time-through", prepro¢essing step Thi~ i~ done this wayinoe the~"pointer" information i~ 901ely a function of a gi~en sit-'~ ~Lwa8 ~ne~o~-ter networ~ geo~etry-which does~'t~change .

,:

2 1~ 9 ~

The prepro¢~ssing, location ~pecific table data generation i~ initiated ~t step 1201 whQre the anemometar location ~alues are ratrieYed fro~ ~mory ana, at ~tep 1202 the ~ite adaptable parameters needed to modify the ~alculation~ are al~o retrieved from memory. ~t ~tep 1203, ~ grid is ¢reated by computing the number of gxid poi~ts in an x ~n~ y Cart~sia~
coordinate ~et of di~en^~ion ba~ed OE the nu~ber of input d~t~ point3 to cre~te a minimal 8i8e ~y gri~ to p~rform the eomputatio~s. At step 120~, ~ set of grid pointer~ i~ produaed to ~ap ths divergenc0 e~timate~
that ~re above a thre~hola ~alue ~rith tha particular poi~ts in the~grid ~y~tem created st stsp 1203. Thi3 i~ to locate the center of ~ microbur~t that ~oul~ be cau~ing an alar~. 8ince a ~umber of grid point~ are ~: ~bove the ~ivergen¢e ela~ent thre~hold ~lue it i8 ~: aifficult to denote the location ~here the microbur~t to be ¢enter-d which ~ould c~u~e tbe~s element~ to :creat~ the Rlarm. ~Each ~8~-40r or net~or~ element is ~ ~ 20~ t-~t~d by placi~g a mathematioal microbur~t ~t each : : gri;d point ~d aoh one of the grid point~ ~o teste~
that would cau~e:~t~ given ~et~or~ eleme~t to bè a~
~larm ~tatu~ the~ a~ oci~te~ ~ith that particular :~etwork el~me~t~.: A re~ult, a set of gr~ poi~t~
:25 :a8 o¢iated ~ith e~oh Lo~ Level Wi~d ~he~r alert 8ystem lOI~:~triangle~ ~na~: e~ga i~ pro~uce~ to :cre~te the lement grid point poi~ter~. In order to p~r~orm thi~
: calcul~tio~,~a~sy~metrical microbur~t model i~ u~d:
; ~ a simpli~tic half ~in~ N~e ~od~l ~hich i~ time i~ari~nt ~n~ymmetria i~ ~oth spnce a~ magnitu~
~na i3 o~ly a:~function of ~mplitude ~n~ ~ maximum : radius. EYe~ though a real microburst ~y be ~patially sym~etriGal, it ~a~ be ~ppro~i~ate~ by : linear sup~rpo~ition of a number of ~ymmetrical WO94/0831~ PCT/US93/09128 2:1~45~3 ~. :

~ierobur~t~ ~t least to a ~irst order ~atham~ti~
e~pa~ion ~hieh produee~ su~ficient speeificity for thi~ ealeul~tion proee~
Onee the above ~tep~ have bee~ performed, the proce8 ing of ~Ba~urem2nt data begin3 at ~tep 1~05, ~here the Low hevel wi~a 8hear Al~rt ~ystem tri~ngle and edge diverge~ee values are usea to generate the : eorre~po~di~g ~ets of ratios of the di~ergQ~e v~lue~
to the thr~shold~, esti~ted 10~8 ~alu~ alarm ~tstu~. A~soeiatea ~ith these grid point~ ~re two s-t~ of magnitude ~alue~: the low level wind shear alert sy~tem diverge~ee to thre~hold ratio~ 2nd as~oeiated~ estimated lo ~ value~. The purpo~e of these two sets of~ ~agnitude informatioh lie in the faet that, although the measured qua~tity i~ wind-faelA divergenoe ~or ~i~d~hear1, the reguired output : ~: ~a}ue~to:;th~ user~ runwsy-oriQ~t~d lo~ value.
~ence a m~pping~fro~ divergenc~ to lo~s i9 needea.
Th~ following data proce~ing steps ~re done ~t e~ch~update sf i~formation from the ~WA3 ~y~tem:
1. I~put~of triangla and Qage diverge~cG
alu~ ~ro~ ~LwAs 8y8te~.
: 2. Com:p~ut atio ~ o ~ ~'r~ti o~"
divergence/thrQshold) for each 25~ triangle:~d edge.
: 3. : Mapping o~ triangle ~nd edge ratio~ to gri~
. 1u~ping of grid points.
5. ~hape:~:gen~ration from clump~.
, Clu p 6ener~ ion ~heorv Figure:2 illu trat~s in flow ai~gr~ form the : clu~p generation p~0033S 111 ~hich rec~ives ~lgoritbm : products from the LOW L~VQ~ 8~e~r Al~rt 8y~te~
: 101 to produce: ~n i~icatio~ of th~ location of wind ::: : ::

;:

~he~r events ~his routine accepts ~s an input the tri~ngle an~-dge ~ivergences producied by the Low ; Level ~in~ ~;8hear Alert 8ystem 101 The clump gener~tion proc-ss 111 then gener~tes clumps of points S that are ~bo~- ~ c-rtai~ inp~t thre~hold level These clumps a~- th-n~ output to the lo~ level ~in~ she~r alort ~y~ste~ ~h~p-~q-ner~tion ~lgorithm 112 The grid po~Dt~ are~ th- data collection points within the pr-~-fined~ ~r-~ ~arou~d the ~irport ~hich ~rea is lo~ pr-~u -~ to be two~dimen~ion~l rect~ngul~r ~rea h~ving s~t of~c~oo~dinat-s in the ~tandard two dimen~ional rectiliD-~r~m~thém~tical orient~tio~ith po~itive x ~alues~to the eas~n~ positive y v~lues to the north The;clu~p~are~g-n-rat-d by fir-t~find1~q grid points 15~ ~th~t~ar-~boYe ~given thr-~hold v~lue t~e~pre-proc--~sing st~ge, ~ gri~ ~ith o s bm 0~5~ m ~p~cing~ ¢onstructea over ~ region which co~-r~ th~ emo eter networ~ 505 A ~imulated ic ~ ~t ~ pl~ced at each grid poi~t and the ;20~ diverg-nc-~i8~ co puter for 0ach networ~ eleme~t If th-~co~p:ut r~aivergenc- for a giv-n -lement i~ above tbat èl~ment~'s ~threshola, an "~ssoci~tion" is m~de betw~ ~en t~e~ grid point and that element I~ this n-r~,~ a~tabl-~ oon~truct-d~that con ~Cted all of 25~; th-;~grid poi~t- to~th- networ~ tri~gl-- ~n~ ~dges ~i~
a hypothetic~l~ diverge~oe detection ~hi~ table is ~ :: : ~ :
.::

:::
::

2124~93 .!~tt~ .

the2 employe~ i~ re~l-ti~ u~ing ~n inYer~e logic.
GiYe~ thak a ~etwork #le~nt dete¢t~ ~ di~erge~ae abov~ its thre~hold, a .~et of grid points (via the tabl~ a~sociated with that ~ivergence, ~ina~ from the theo~t c~l a~aly~i~ the~e points ~re potential microbur~t lo¢ations. ~.
o~oe th~e subsets of grid poi~ts h~ve been id~t~fied, they ~r~ prsces~ea to g~eratQ "elumps~ of . .
~ csntiguous group3 of grid poi~t~. By ~o~tiguous, ~t , :i~ mo~nt t~at ~djace~t up, do~n, right, or left poi~ts ~re con~idered, ~ot tho~e ~long ~he diagon~l. Three sets of clump~ are g~eratad to i~clude gxid point thresbold data repre~e~tati~e of ~low-leYe~ high vel~, anh "low-le~el-de~sity" coll~ctio~ of gri~
~S poi~ts. ~h~ "low-la~ d "high-level~D grid poi~ts axe i~dic~ti~e of the ~g~itude of t~e esti~ated ~i~
:, ~
ivergence at tho e particul~r grid points. The "high-le~el" grid poi~t~ are repreQe~tativ6 of a co~dary thr~shold u~ed to disti~gui~h ths grid : :: :~
2Q point~:th~t~havs si~nifica3tl7 e~ce~e~ the ~ltial ;thr-~hold. ~ This ~econdary thre~hold thereby diff~re~tiat~ wi~d ~hear~ of :sig~ificant ~agnitude ~ro~ tho~e of moderata magnituae.
'Low-l~vel-de~ity" g-id-poi~t clump~ are id~tical to tho~- for thq lo~ level ~nd high le~el proce~ discus~d abova but repxe~ent ~ co~de~atio~

~ ' ' '-.

W094/08~12 212 ~ 5 g 3 PCT/US93/09128 , of a large number of grid points, Which number would be overly large or the re8ultant g~ometric pattern ~oul~ be concave or exten~e~ in nature An es~mple of such a problem ~Oula be a col}ection of grid points that ~re located in a figure eight shape In order to re~uce the collection of grid point~ into 8mall, conVes and ~co~p~ct~ patt~rns, ~ den~ity ~eighing operation i~ performed on the low level grid point values; In~or~er to ac¢omplish this, the original 0 ~ ~gDitude of each grid point is ~ult~plie~ by ~ local n~ighborhood~ occupation density weight to compute ne~magnitud- valùé at each grid point to thereby more accur~t-ly~r co~figur- the g~ometric pattern of the~e ; gri~ poin~8~ ~hQ~de~sity ~eight i~ ~ norm~lized value l5~ b t~--n~zoro~ n~one Yhich i~ g-ner~tR~ by ~ny one of a~ numb-r~of ~thematlc~l metho~ depe~ing upon a given point'~;location in the grid For example, the ~eighborhood~8et~0~ point8~for a given i~terior point ~re the ~ ght;~adjacent point~ inclu~ing the ~iagonal~
20~ and~ th- given~point~it--lf The ~umber of point~ in thi~8et th~t~arè~abo~e a thre~hol~ value ~re ~u~med and thi~ total~Dumber i~ di~i~e~by the n ~ er of grid pointJ thst~ ~r-~in th- origin~l neighborhood ~et The8e den~ity ~eighted point~ ~re then for ed into 25~ clump~ in ~n~i~entic~l r~shio~ ~ for~ the low level ~nd~ high~ le~ co~put tions to form the ~ow }evel :.

WOg4/08312 PCT/U~93/09128 212~93 ~30- ~

dRnsity geometric clump~ This procedure conasnse~ ~;
the collection of grid poi~t~ into ~ore compact patter~s ~nd a1so separatQs overly extended clumps into ~ set of smaller, compact clump~

,:
S Preferr~ 6eo~etrio~8hape A single~;preferre~ geometric shape is use~
throughout the~- co~putatio~s in order to ha~e con istency~a~d;~simplicity of the calculations The . ~
preferred ~hape~ o10 ed herein i~ a semi-rectilinear ~10 oval akin to ;the~ shape of a ba~d-aid, that is a rectangle ~ith ~s~mi-circle "end-caps~' ~these microburst shapes;~are the same as the TDWR shapes) hi~ sh~pe is -athematically defined by an ~xi~ li~e segment~havi~g two~en~ point~ and a raaius u~e~ at 15;~ -ach~of~th-~ ~n~points to~define a ~e~icircle Thi~
q~ometric shape~is produce~;for each clump such that the à~is~1in- segment has th- minimu~ ~eighte~ squared istanc-~from~a11~;of~the qrid points that are withi~
;this given ciump~ an~ furth-rmore, thi~ shape enclosea 20~ a11~of tbe~c1ump~'s~grid point~ I~ case~ ~here the shape is overly~large or conc~ve in nature, the ~hap~

~: : : ~ ,: ~ .. .
; is proce~sed to create a ~umber of smaller shapes vhich enc10~e~t~e~grid points This shap~ i produced such that it is of ~i~imum area after sati~fying these ~ ~2~5 ~ conditio~s A~ further processing step, a lea~t-: ~ ~ ,,';

:

~ .
,`, WO~4/08312 PCT/US93/09128 212~93 ~quare~ ~ize reduction, i~ the~ performed to "txim'~ ~
overly l~rge ~h~pes. I~ co~puting the ~hapes for ~-microbursta, the magnitude i~ormatio~ u~ed is thQ ~:
r~tio of the calculated ~ivQrge~oe to the t~reshold that i~ ~apped from triangle~ ahd edg~ into the grid po~ts. A giv0n gri~ poi~t'~ ratio ~alu~ i~ ge~erated : ag follo~ r~t, a r~tio or ezch ~WA~ net~or~
:.
~eleme~t: ttrla~gle:and~or edg~ ao~pute~. This : ratio i~ the ratio of that ela~e~t~' d~te¢te~
~:~ 10divergeace stimat- a~d t~at el~ment~' ~ivQrgenc~
: threshola value. Thi~ predetermined thre~hold i~
designed to i~dicate h2zaraous wind-fi~l~ divergence, : ~ comput~ ba~ed~: upon a ~3thematic~1 ~icrobur~t si~mul~tion, ~and take~ into aecou~t the geometrical : S ~ture of the~gi~e~triangl~ or edgs~ anot~r set of :
:;: ~ag~itu~e i~form~tion ussd i~ an ~R~ociated lo~ v~lue o~timate~or~e-oh:poi~t, ~a~ed o~ the~e di~ergence~.
he ~ioroburst ~hape are calculate~ ~t the "wi~d ::: : :
h-~r ~ler~ 8A) level using t~e:low level a~ity ~:clu~ps,:le~t~gu~re~ sh~pe ize reductio~ ~a the tati3tical ~h~pe~ ma~itude ¢v~p~tstio~. ~he ot~er et o~ geo~etric ~ape~ i~ at th~ "~icrobur~t alert"
~MBA) l~el~u~i~g tha high le~el clu~p~, le~t s~uaxeY
reduction : and~ the ma~imu~ valu~ of ~ag~itude :
comput~tio~
; ~-. .

.

W094/08~12 PCT/US93/09128 212~9~

Clump Ge~eration Proce~
Figure 2 illu~trate~ i~ detailed flo~ diagra~ the clump generation proce~Q 111 which proces~
i~itiated ~t step 1102 ~here the data is received from the as~ociated low level ~i~d shear alert sy~tem 101 and ~tored in memo n. At ~tep 1102~ thQ clump gener~tio~ proce~s ~ 111 co~vert3 the low level magnitude points iDto 106al occupied ~eighbor density weighted magnitude value~0 Thi~ proce~ a~ discus~ed .
, ~ above use~ all:~of th- low~level input magnitude value~
: ~nd co~pute~ew values for the e poi~t ba~e4 on the de~sity of ~diaceDt dat~ point~ th~t h~ve e~ceeded the initial: predetermi~ad threshold. ~ach give~ data point:that is above the~i~put thres~old ~lue is giv~n 5~ ~ deD~ity weight~ ~hioh is a nu~ber betwe~n zero ~d ; one indic~tive~of the Dum~er of contiguous grid point3, inclùdiDg the givon poi~t that ~re above the input thre hol~val~e, divi~e~ by the total number of contiguou~poi~t3. That i~, for a~ intRrlor point the 20~: de~s~ity~weight~is the ~umber of ~eighbori~g point~
~bove the~input~thre~hold ~lue: di~i~ea by ni~e. This beoau3e :th- ooDtiguouq poiDts i~ defined ~s the djacent points to the left, r~ght, ~p, down ~a the four diago~al poi~t~ i~ thiQ xy Cartesi~ coer~i~ate ~ ~.
2;5~ system. ODCe this set of density:~eighte~ values hn~e been computed, processi~g aa~anoes to ~tep 110~ :~

` " 2 1 2 ~

~herein the initi~l gsouping~ of d~ta point~ i~
~ccompli~hed by groupi~g the grid points that have exceeded the threshold ~alue into contiguou~
groupings Concurr-ntly ~ith the operation~ on low level density data point~, or subseque~t thereto, the steps llOS an~ 1106~ are e~eouted o~ the high levél agn$~tud- points~ to perform the s~nle co~tiguous grouping function of 3teps 1102 ~nd 1103 The ~et of groupings i~ then~use~d at step ~06 by the shape 0 ~driver~to~generate~the predetermine~ geometric shape of mini~um ~rea ~
Using~poi'nt~ that are ~till i~side the ~hape art-r~ radiu~ r-duction compute the lea t ~quale~
reducèd~;~Yi~ ~egmènt to produce a ~ew reduced aYiq 15~ ~line~s-gm nt, ~he xesult~nt reduced shape a~i~ line egment i~ then ;con~erted into the original, non-rotatéd~C~rté~n~coordinat~ ~y~tem ~nd the overall ~gnitu,ds ~for~tb-~shape is computed The resultant sh~p-~consi~t~of~ lin~ ~hose' e~ poi~ts repre~ent 20~ the;c-nter;~of~ sem~oirole of~pr-determined ra~ius *,hich~e~d point~semi¢irole~ when oonn~cted by ~traight line ~eg~ent~create ~ band-ai~ ~hape to e~clos~ all of the data~points in a m nim~l r-- ~hose m~gnitude has b n~calculat-d Bimilar proc-osing of the input ~ data ta~es place~for the high level~magnitude point~
in steps 1106~nd~ 1107 the proce~sing of ~hich can :~: :: :
, ~ ~

WO 94/08312 PCr/US93/0~128 212~93 ocaur ~equentially or in parallel with the operatioa of s~p~ 110~, 2~d llOS. o~ae the shap~s ~d th~r magnitude have been calculatsd for both the low level dQn3ity magnitud~ points a~d the high level mag~itu~Q
5 point~ proce~3ing exit~ at step ~109.

8haPe Production As not~d abo~re, thi~ predetermin~d geo~etric hape i~ a bzu~d-aid~hape ~hich i~ ds~ined by ~n ~Ci8 e ~e~ent having t~o e~d point~ ~d a radius used .:
0 at the end ~poi~ts to produce two ~ ircul~r ~hapes.
Thi~ proce~s ~is ~ illu~trated in flow d,iagram fomt in Figure ~. The: proce~s ~ initiated ~y retrieving all :; of the gri~ poiDt~ i~ one of the a~o~re ~ote~ s~ts stori~g th~e;: i~ ~em~ry. 11 ing the~ ~tore;lS gri~
15 point~, the ~-a ured or calculated nag3~itude of e~ch grid point in~ clu~p i~:nor3lalized. O~c~ all o~ the grid point valu~ i~ the ~et have b~en ~ rmalized, ~

; , . ~
~oight~d lea~t ~quar-~ line is fit t~rough these point~ usi~g ~ a st~dard weighted lea~t ~guar~
2 0 ~ ~techniqu~ . ~his produces th~ ~est li~e fit through all ol~. the ~lid point~ i~ t~s i~put ~et of gria points. Onoe tha Y~ighted lQast ~gu2re~ li~e ha3 been prc)duoed, the e~ds of this li~e ~g~e~t are c~ ulated by proj ecting ~ll o~ the data poi~t~ in the ~et onto 25 the computed lea23t ~quare~ line. ~he proc~ U~eQ the :

WO 94/0$312 2 1 2 ~ S 9 ~ Pcr/usg3/ogl28 .....

coordinatss of e~:h of th~ data point~ a~d tb,e ~lop~3 o~ the c:omputed lea~t ~quareq li~e through the~e poi~ts. The ¢oordinate-~ o~ the clump poi~tQ ar~ put into a rotated s:oordinate ~y~tem ~uch that the le~t 5 ~quare~ line i~ hori~ontal. The output from this c~l¢ulation i~ th~ clu~p point coor~ alte3 i~ thi~
rotate~ ~y~te~ an~l t~e a~i line ~egDIe~t eDd poi~t~
al50 i~ thi~ coordi~ats sy~tQm. Th~ ir~t set of coordinate ~a~lue~ o~ thi3 rotated e~d point i~ the 10 leftmo~t point C~D the lîne repre~ent~ti~e of the ~allQst x ~alue in the rotated xy Cartesian coordinate y~tem ~d the se~o~d coor~inate output i~
the riq~tmost poi`nt represent~tive of th~ l~rge~t x :~ v~lue in thi~ C~rte~ian ~oordi~ate system. O~¢e the ~nds o~ the sh~ap~:lins ~egment have been d~ermi~ed all of the subse~ue~t Gomputations are don~ i~ the , rotated~:coordi~ate 9y~tem. The r~dius of ths ~ape that eaclos~ h~ poi~t~ ~nd i~ of ~i~i~u~ ~rea i~ :~
cslcu~atQ~ by ~i~g a o~e ~im~ io~ mooth-fu~ction, ~:
20 ~ ~i.e., mo~oto~i¢) mi~ tion routi~e. -~
, ~ -Bhape Area ~i~i~i~atio~
~: ' . :.
The minimiz3tio~ function i3 then acti~te~ to compute th~ radius that minimizes the shape are~ ~d u~i~g thi~ ~ew:radiu-Q a r~view i~ mad~ to determi~e ~hsthex the ~xi~ linQ ~egment e~ poi~t~ ca~ be . ,.
:

WO94/08312 PCT/US93~09128 2125~93 -36- ` `

modified in vie~ of the deter~ine~ ra~ius This i9 aone by proje~ting the ~alid aat~ points in the current set onto the~computed least ~quareQ line and computing new end~points as discussQd above once s this i8 done, the ~Yis length i~ reduced if possible by mo~ing the ~Yi8 end points to~ras the ~i~ seg~ent bary center using a ~eighte~ least square~ re~uct~on of the horizontal ~ist~nce from clump points to the closest shape boundary By closest, it i~ meant that 0 ~ tbese points are~partitionea into three Qets a set whose x values re less than the shape_ bary center, a~s-t ~hose x values~-r- gr--t-r th~ the shapes b~ry ¢-nt-r ~na~ t~of points that ~ere originally a~sooiated with~the~Qhapo but ~fter radius reduction 15~ ~ar- ~ow~outsido~t~he~shape The norm~lized ~eights are s-1eot-a to~be~unotion of point~ magnitude ~nd its ; distance to~the~is segment bary center The proces~
uses~ th-;current`; w c-ss lin- s-gment end point~ ~nd ¢o~putes th-~b~ry~ center ~of the curre~t ~is line ;20~ ~egme~t~ ~nd~initi~ es the minimization iteration If the ~9h~pe~ ~90 gen-rated is too l~rge, it is dis~ected into a plurality of shapes The te~t of ~Oe991~Q 9ize~ i8 that the 1ength of the axi8 line 25~ ~ segm~nt~ pIb9 ~ twice the radius is greater th~n a predeterminea threshola If so, the ~'8 li~e segment : ~ .
.

W094~08312 PCT/USg3/09128 ~ ` 2124593 -37 i~ divided into 3~aller and pota~ti~lly o~erl ppi~g pieoasO The grid data poi~t~ origi~al}y a~ociated ~ith th~ original clump are then as~ociated with the corre~pondi~g ub~hapeQ. If there i~ an overlap of the multiple shapQs, the grid data poi~t~ can be as~ociated ~ith more tha~ one sh~pe. The resulta~t pluraIity of ~h~pe~ more ~ccurately reflect the ~co~current e~ist-no- of ~ultiple adjace~t or o~erlapping wind ~hear e~e~ts.

Lea~t 8qu~res 8hape~8ise Reductio~
~. :
Thi~ proc~s3: pro~iaes for a ~impl~, efficie~t a~a mathem~tically~ rigorou~ m-t~o~ for ~ore preci~ely ;~
iadicating the ha~rdou~ ~icrobur~t r~gion. The :.:
origin~l Loroburst shap-~ algorithm - still uset i~
lS~; ~the; TD~R system,~ require~ th~t all of th~ ~h-ar- -s~egme~ 80~t~tb-~"run~ of radi~l ve~lo~ity i~creas~") be::~e~closed:~within the ~ieroburst ~hape~s) ~03.
Fi ~ r-~8) I~f;the locus of th~s~ ~he~r ~egme~ts 80 is~ ove:ny ~ ~e~tendea ~n~/or f~irly con¢avo in ~ 20~ geom-trical structur-, the "all ~nclo~i~q" shape 803 :~ can be toQ larg~. ~hat i~, it may contai~ ~on- :~
: hazardou~ regions 805.: Thi~ can generate falqe ~l~rm ~., warni~gs as a~;runway alarm is~ genor~ted when ~y portion o~ a microbur~t~hap- 803 int~r~ect~ ~ pr~-defin-d box 802 round ~ giv-~ run~y 801. This sam-;
':

,, WO94/08312 21~ 4 5 9 3 PCT/US93/091l8 situation ~pplied with the LL~ ~icrobur~t ~hapQs ~here herein, w- ~r- ¢oncerned ~ith o~erly extonde~
and/or conc~v- g~rid point clumps, ~ oppo~ed ~hear-sQgment clusters, though the concept is identical S Th- ~olution to this documented "overwsrning" problem has boen de~eloped in the conte~t of the le~t-squ~re~
re~u¢tion of the Qbape-size for the LL~A8 microburst shapes in th-~apparatus of t~e present invention .:
A further contributio~ of the ~overwarning~
10 ~ problem, i~ in~th-~ gen-ration of the "magnitude~ of the run~y ~lert That i~, after ~ given microburst hap- ~803 ~int-rs-¢ts ~ "r~n~y ~lert-box" 802, a mag~itu~e for~the~alert ~UQt be c~puted Agai~, the tec~niqu-~us-d~for~the~ TD~R stan~-~lone system i8 15~ fairly~ ~implistic and t~n~ to over-estimate the haz~rd~m~g~itud- ~ The~- over-e~timate~ ~re often viewed~a~fal~--al`arm~ by the pilot~ Therefor-, again~in~the~oont-~t of the LLWA9 miarobur~t h~pe~
a~ simple, ~-ffi¢i-Dt ~na mathem~tically rigorou~
20~ methoaoI*gy~i~ u~-d in the ~pp~ratus of the present i~vention ~ Thia~ algorithm employs a stati~tical ésti~e for ~`gi~n microburst ~h~pe's m~gnitude A shap- i~ d-fin-d by two aYis end point~ ~Sel~

Ye1) ~d ~Ce2~ Ye2~ ~ [a~ e2] ~a ~ radiuQ R (Figure 2~5 ~ 7) The ~hape ~ i9 generated initially by finding the line wSich, in~ ~ least square~- sen~e, (weighted by W094~08312 PCT/US93tO912~
212~593 mag~itud~ be~t fit~ the set o~ poi~t~ in a gi~e~
"alu~p~ he~e ~lu~p poi~t-~ e3~entiall~ re~lect the ~iv~rgence msgnitu~e at those poi~ts i~ ~pace -~stimated from the LL~A8 ~ind fiald.
~he radius is the~ fou~d by ~n iterati~e procedure whi~h minimizes the area ef the ~hape whila simultaneou~ly reguiri~g that all poi~t~ in t~e clump : are enclo~Qd. ~his technique i~ ide~ti~l to the proaedur~ u~ed for TDWR, ~hi~h u~es "~egment .
ondpoi~t~" ~9 opposed to "points i~ a clump". ~e~t, . ;

we try to reduce the ~hape 8ize ~0 that it gi~es ~ .
.
etter fit to the~poi~ts. This i~ do~e be~au~e the : origin~l criteria~that ~11 point~ be enclo~e~, t~n~s .. . .
::: to result in :o~erly-l~rge ~hapo~ whe~ th~ olump i~
fairly: co~a~-. ~; A further u~desir0~ aomplication o~curs bsc~us~ of the g~nerally "woaker-~gnitud~"
point~ on~ th~ ~dgs~ o~f the clu~p. Thi~ c~n be conceptualized by con~idering a ~ ~ etrical ~icrobur~t outflow. Th~ clu~p po;~t can be Yi~ed a~ ae~cr~bi~g 2~0~ ~ coatour-le~ ; of diverge~G~. The ~ ~ter" of the olump b-ing~ th- "5ent-r" of th- microburst outflo~.
Th~ highe3t 1~-1 of divergence wouId be at th~ ter of the microbur~ outflow, the~ monotoniGally d~crea~ing in magnitude with incre~ing di3t~¢e from ::
the cent~r. ~he ~hape's radius i~ first reduced, then : ~ : tho a~i~ length~ Both ar~ ~on~ u~in~ a ~ight~ laa~t ~quares t~oh~ique.

Reductio~ of the 8hnDe ~adiu~
What we do here i~ reduce the ~weighted) di~tance of the (originally) -nclose~ points, (2kp Yk), to the -qh~pe bound~ry.:
We:have:tbat R = dk + dk, where R is the original radius, dk i9~: tbe perpe~dicular ai~ta~ce from the point-to the shape:~xi~ ~or axi~ endpoi~t if Xk S
Xe~, or Sk ~X,2~, and dk i~ the distance from the point to the bou~dary.
Therefore~w- minimize dk - R - dk~ which le~
to th~ ~eight~d le~t squares equatio~ for R, the now ~;radius:

~which has the so~utio~:

when we choo~e a ~et o~ normalized weights ~k ~ ~ Wk wo g4/083l2 2 1 2 4 ~ 9 ~ PCT/US93/0912~
, . , ~

~e ~efine the ~eight~ to be:
m~d~

where mk i~ tho gi~en ~agnitude ~t each poi~t. Thi~
~eighi~g i3 uS-a to remo~e the bi~s ge~eratea by the relati~e higher de~i~y of the i~ter~al poi~t~. Thi~

, can be under~tood by con~ideri~g a ~h~pe ~hich i~
~ - .
:: : : disc, ~n~ ~whose ~o~stituent ~lump-point~ all ha~e equal magnitudeq. : If the ~eighi~g function only con~i~ere~:~ag~itudes, then the l~ast 3g~are~ ra~iu~
reauction ~ould~a1wa~s attempt to ~ake ~ ~ew ~isc of o ~:~i~imal-ra~iu~. The u~e of the ai8tan49 ~alueg i~ the ~: ~eigbi~s functio~ designed to cou~ter~ct this tende~cy. Furthe~or~, ~e choo~e ~ coor~nat~ ~ystem rotated~su~h;~that~the ~ horizo~tal.

* i~dicati~g rotatsd ~ooral~tes) ::qsIn t~io c:oordin~t~ ~yst~, the dk'~ aro gi~e~ byO

:: :

(Y.~ ; ~<~,~
d~ ¦~ -Y~
~[(~ (Y~ J~ ~>

: ':

., : ~

W094/OB312 ~ PCT/VS93/09128 ~,~
- --42-- r Reduction of th~ _aDe ~i8_ Length Next, ~e r~duce the ~i~ length by ~separ~t~ly) moYi~g the axi9 8egmQnt endpointQ towar~ the seg~t --mid-point. ~e u3e a least ~quare~ reduction of the ~ .
: Shoriæont~l ~in rot~ted ~oordin t~ di3t~¢e from ~ i`
g~en point to the (cl~e~t) bouhdary. Note: the axi9 : ~ i reducea only~when the axi~ length i~ lv~ger than ~
hr~shol~ le~gth (4pproxi~ately 1 k~). By "closest", -.
~e mean th~t~the cl~p points ~re p~rtitio~ed into 10 three ~et~: a ~et ~ho~e X-~oor~nate3 ~re le~ than thei~h~pe ~ eg~ont'~ mid-pointr X; one "great~r~
than" X;: ~nd a third ~et consisti~g of tho~ point~
that ~after radiù~reduction) are out~ide the shapa~
do not ~e~this t~ird 3et of poi~t~ ~i~¢e their IS ~ hori~o~tal) ~i8~t~n4e to the ~ou~d~ry i~ ~ow) undefinQa.

~:
-,: : ~ ' ` ~
Therefore,~the proble~ we ~re ~ryiag to ~ol~e ~for ~ generic endpoi~t "e") i~

; ~ tr=d~-(~o~Xo) ~ ~ where dk is:the horizo~tal (~ ) dist~nce fro~ poi~t :~ : ::: : :

~ ;
.

WO 94/0~312 PCI/U~93/09128 ,~.'.t~ ~12~5~

to the bcsu~d~ k i~ ventual~ lea3t :~quare~
ais~ta~:e; -~e a~ld ~e ar~ ~i~ilarly the origi~al ~d ls~st ~quares e3l~points.

The ~ ~ndpoint W8 ~8~1t i~: ~
: ,,' X,=~W1~d,~

5 ~her~ t~ ~et of point~ j ref er~ to eithl3r point~
grszlt~r t~n X ~or the "right" ~D.dpoint c~r les~ than X for th~ "left~' ~dpoi~lt, respecti~r31y. Th~ ~reights ara ¢-ho~ to be:

w~= m~ 1$-~1 -wh~re: ;

W~

0 ~ ~ A~:be~or~, th-3;~eight~ ar~ cho~en to reduc~ over-bi~
by poi~t~ alo Q to a~. .

-. .

WO ~4/08312 P~/lJSg3/09128 -~4- :

The horizo~tal ta~)-di~tanc:e to the boundary d gi~e~ by d~
= (Ra_y ''~

:
:
,~

~he value ~e ~rant to ~i3limize i~ the~: .
:

:

e I.. i~ th~ hosi~ontal ~ t~e ~roDI the poi~t J
5 6~, YJ'J to the~lea~t ~quara~ r~du~aa ~ou~aary, a~l ~Sj ic the horizo~tal di~ta~ce b~e~ S~ ~nd 2e:
:
~=(X~ ~)lJ2 R i~ t~e ~lea~t sguaro~ reduaQ~ ra~iu . ) .

``;

W094/08312 2 1 2 ~ PCT/US93/09128 T~r~ore, the n~w ~a~point, ~e i~ gi~e~ ~y (aga~n i~ rot~t-d ¢oordi~ntes3 PIJL(R~-Y;~ ~;]
.,',.
~h~re:

Note the Qam- va1u~s re~ult ~or poi~t~ between X an~
~Xe, and ~ a~d~the bou~dary Furthermore, the ~me - ~ r~ult applie~ to~point~ o~ either ~id~ of X That , the~s~- qu~tio~3 apply equally for ~oth s-t~ of poiat~ pa~titi~on-d ba~ed upo~;beiag less~tha~ or gr ~ter-tha~ X) ;I0 ~ ll~8~icro~ur ~ h~pes ~cDi~ud-_~g~pytatio~
his~routiaa ~o~put-~ ~n ovora11 m~gnitude tim~t~ for -~ g~-D 3hape T~ ;t~chniqua is to ~ -~ 8t~de~t'~ t-3tatistic ~istribution for khe m~gni~tud 3~for~the ~-t of poi~t3 associsted with th~
l5 ~ Qhape~ Th~ 3hape ~ag~itude i~ then the perceDtile ~alue give~ by~the ~ea~ ~agnitud~ plu~ tand~rd de~i~tio~3~This i3 an ~pplio~tio~ o~ the ~ell-X~own co~f~d~n~e~ -nt~xYal" tech~que ~fro~ ~tati~tic~l thaory ~ Thi~; di~tributio~ ~9 chose~ for its :: : : : : : :~

. . .
.
' '~

W094t~8312 2 1 2 ~ S 9 3 PCT/USg3/09128 ~

-~6 i :

applicability to ~mall sa~ple ~et~ ~nd its approxi~ation to ~ normal ~i~tri~utio~ for ~a~ple ~ets of arou~d thirty element~ or more. ~urthermore, the ~lue of "~" th~t ~8 bee~ u~ed (~ = 1.3), wa cho~n to appro~i~ate ~n 80 to 90th peroentile ~alue over wi~e range of ~agrees of free~om, twhich i~ the ~umber ~ of poi~ts mi~us on~

; ~ 8~et~y_~e~t ~ymmetry te~t ~pparatu~ 113 validates th~
o microburst 8hap-3 produce~ by microburst shape~
genor~tor 112 ba~ed o~ the ~u~il iary ~nformat~on pro~uced by the: ~e~tura~ aloft ~ sh~pe i~orm~tion ob:tained ~rom ~t~e ~er~ina~ Doppl~r We~ther R~d~r ~8y tem 502. Thi~ vali~tion ~etermine~ if thare i~
5~ supporting avidenoo that a gi~e~ ~a8 microbur~t hape, i~ truly ~o¢iated with a ~icro~ur~t. That i8 ,: tho ~hape~that i:~ gen~rat~d from the deteetion o~ :
surfac:e ~ fiel~l ~i~er~c~ caJl be assoc~n~ ~ith ither ~ micro~urst or ~omo other type of ~iad fi~l~
20~ ~no~aly,:~u~h ~ th~r~al ~ctivity, ~oisy ~in~ , etc.
~ince ~g~etry ~8umptio33 ar~ implicit in a g~er~tio~ of mi¢ro~rst shap~ a~ these ~ umptions .
are ba~od o~ th~ a~ociatioA of the ~urfac~ rgance iith the microburst. In ~on-microburst ~ituationQ, th~e a~sumption~ c~n :le~d to the generation of WO g4/0831~ PCI`/US93/09128 2124~93 _"7_ unwa~toa fal~e ~his ~meltry t~t pro~:edure 1~.3 re~ovo~ theu~w~ted al~s ~y re~ie~ri~g reflecti~ity a~ld microburst pre~:ursor i~formatio~ fro~
the q!~ l Doppl~r ~eath~r Radar ~y~t~ 502. The~e 5 i~put~ are coDlbia-~ ~p~tially a~a tempor~lly to fon~
~etry disk~ ~rho~e pre~en¢e i~aic~t~s the pos~ible ~ e:cist~ce of a microbur:~t w~thin or ~bove it~
boundary. ~he~ gi~-n microbux~t ~hape that i~ to be ::: validat-d by the ~ymmetry test 113 is t~en te~ted for : 10 it~ pro~i~ity:to a~s~etry di~k. Ther~or~ ak microbur t ~h~pe that i~ close to a ~y~metry ~isk is : ~ valid~tea:~nd t~ose that ~re aot ~re pr~umed to be , erron-ous a-t-ction~
Tbi~ ~--tn t~8t 113 iB in't ti~t01~ ~t ~t~!p 130~.
: 15 ~ th retrie~l~o;~ite Qpecific para~eter~ ~rom memory to : o~ify: th-~ ~ oal¢ul~tio~ ~a~ed oa loc~l clim~tologic~l~co~itio~s n~ ~en~or co~figur~tion.
: : : At st~p 1302,~a rect~ngular gri~ i~ the sy Cart~Bia~
coorainat- ~stem;~iJ producQd co~ ti~g of ~ imal 20 Bize gri~ es~ry to an~lyze the c~lcul~te~ ~hape~.
At step~l303 the microbur~t ~h~pe~ ~re ~le¢te~ whose magDitud- ~re -gynl to or gra~t-r th~n a site ~aptable thre~h~ld. At ~tep 1304 th~ pr~se~t grid poi~t v~lue~ ~r- computed ~ ed on current Terminal Doppl-r Weath~-r ~ad r fe~tur~ ~loft in~o~mati~ ~nd : aay T~rmin~l DopplQr Weather Raa~r or Lo~ ~e~el ~ind ,", W094/08312 PC~/US93/Q912~
.21~5~3 -48- .

8hear Alart ~ystem ~i~roburAQt ~h~pee. The fQaturQs ~loft i~puts ~re i~ the form of di~ sscribe~ by xy center coor~i~ate, ~ ra~ius, ~ ~ type: low r~flectivity, ~torm ~ell, refl~ctivity oore ox : 5 microbur~t precursor di~s~ A ~agnitude value for ea¢h of the~e fe~ture~ aloft ~ s i~ a~signe~ ba~e~
upo~ it~ type. Th~ ~icrobur~t shapes herein are those that hav~ been filt-r-d out previou~ to thi~ routi~e a excee~ the predetermin~d thre~hol~ ~alu~s.
Therefore, all of the Low Level Wi~d Bhe~r Alert 8y~tem ~nd Terminal Doppler Weather Radar shapes ~ comput-d ar- scr-ene~ to come up ~ith A ~omposite ~t : ~ of s~apes that e~cee~ a given thr~hold ~alue. For :: :
-ach dis~ that i~p-ct~ th- ~nalysis grid th-t ha~ be~
: 15 produc~, spocific:gri~ point~ ~ithin that di~k ha~e : their ~ag~tuae up~te~ b ed on the nat~r~ of the ; dis~. ~¢h :~ qrid~ point m~g~itude ~lue is time filter~a ~ith~a;: ~ingle pole recursiYe f:lter to enforo- ~ ~en~-~ o~ ~tim- co~ti~uity. This 8~ of 20 ~ filt-r-d m~gnitud~9 is tho~ t~e output of thi~ routi~e : to ;tie cr~ate:-~mmetry disks st~p 1305. The dis~
magnitude~ ~re ~-lecte~ by appropri~t~l~ ¢hoosing base or minimal ~alue~ for eac~ input SQt ~0 th~t the : ~
feature~ aloft~disk typ- rel-te~ to the:value of tha Z5:~ :: actual los- ~agnitu~- O~ce thes- gri~ v~lue~ haYQ
b--~ ~t~bli~b~ at ~top 1305;~the ~ymmetry ai~k~ are :.

', , ~:.

W~ 94/08312 ~ PCr/U~3/09128 _~9_ cre~te~ u~ing a slightly modifie~ ~er~ion of the clu~p ~d ~h~p~ g~ra1;~o~l algorithm discu3sea ~ov~. O~:e the~e ~hQe~ haYe bee~ creat~d at ~tep 1305, at step 1306 th~ s~etry te~t i perfo~:med to ~alidat~ the 5 we~ker I,o~ I,e~-l lri~d hear Alert ~y~t~m micro~ur~t shape~. The I~L~a8 microbur~t ~hape~ ~d ~ymmetry di~ks ~re tbe i~put to this ~tep ~ ~y Lo~ I Q rel Win~
8hear AleEt ~y~te;~ microburst ~h~p~ ~ho~ g~ tuae ~
e~l to or abov- ~ thre~hold ~alue automatically 10 ~ pa~ses the test. Oth~ , a c:ircum~cribing dislc i~
cr~ated ~round each of thess ~e;~k shap~ ~d a te~t i8 perforDIed to see whether a giYe~ Iiow Le~d ~i~d 8h~ar Alert ~ ~yst~ dislc i8 close to a~y sy~metry di2~k. If it i~,~ th-~ that Low L~el ~in~ ~hear ~l~rt ~te~
- .
~ ~h~p~ p~s~ the~ test. !r~e outpt~t of this prooe~ is -::
a liat of logioal value~ for each of the ~ nput I.ow evel~ 8he~r Alert 8yqtem microbur-~t ~hape~ ~o indi~te re~lts of thi~ syD~Qt~y t~st ~ith a trua ralue i~dicatillg t~at the~ ~ap~ h~ pas~ad th~ test ;:
20~ ~;d is ~a valia f or usa i~ cre~ti3~g æ microbur t ~l~rt.

~iorob~lr~t Intearllt~ o~
. .
: ~ Tho ~icroburst integxatio~ apparatuY 114 i~ the driv~r of tho microbur~t portion of the i~tegration appa~atu3 . ~ Thi~ ~pparatu~ con~ert~ thg~ ~ermi~al Doppler W~athor Radar microbur~t ~hape~ and vali~ated ;:

WO ~4/08312 PCI`/US93/09128 Dlis::roburst ~hapes output by ~ym~stry tB~t app~r~tu~
~13 ~a t~ Low Level ~ a Bhe~r Alert ~y~
roburst ~hape~ ltO ruaw~y specific ~lert~ for ~y region~s Oll the operatio~l runways ~ ~rrival R1, aep2lrture Rlo otc.) that are defi~ for the physic~l ru~ray~ Rl-R4 in the a~oci~t~ pre~ete~mined are~
which are ~f f ected by the shapes . The regions so af f ect~d are ~ombine~l with the Low Level Wi~d 8hear Alert ~y~tem run~r2ly orie~t~a 1oS8 alarm3. The I,o~
I.evel ~ind 8hear A10rt 8ystem inputs to thi~
~icroburst integratioll apparatu~ 1~4 are the rlmway ox:iente~ Os~es th~t are th~ output~ proauc:~ by the I.ow Le~re1 Win~ ~he~r Alert ~y~teDI 101. The ~ioroburst i~tegr~tion apparatu~ produces array~ co~tai~ling the ~agnituae a~ loCZ~tioB of a~y 1053 al2lrm a~ mappe~
onto the r~ay oonfigur~tion ~rithi~ th~ pred0temli~e~
: ~ro~. Th~ D~i¢so):)urst i~tegratio~ npparatus 114 rec~ives q!~ a1 Doppler ~leather Radar ~icroburst hap~s from the~Termi~al Doppler ~lleather J~aaar qy~t~m 20~ 502 ~d co~Qrt~ th~ by laappi~g th~ into ru~w~y : ~ ~ ; specifiG locus ~ a~it~ dications~ to pæoduce ru~way a1an~s. In a~ditio~, miarobur~t ~ape:s that ~re computQ~ ~rom the I-ow Le~re1 Wi~ ~hear a1ert ~y~te~ 101 a~ ~lidate~ ~y the symmetry t~t a~pparatus 113 ~re Z1130 c:o~varted into ru~way ~larDI~ o~ce they Im~e suf f ici-~t ~aagn~tude or the ~mmetry hypoth~

WQ 94/08312 P~/US93/09128 2124~93 of Qy~metry ts~t app~ratu~ l~ 3 sub~t~nti~te~ t~e~ r e~ te~c:e. In ~it~ o~" a~y Low L~vel ~i~d ~he~r Alert 8ystem rullway orieat~a 10~8e~, 213 produced by I,ow I.e~el Wi~ 8hear AlQrt ~y~tem lOl, th~t ~re 5 co~curre~t with ~y I o~ I.9vel ~i~d 8he2~r Alert microbur~t ~hape~ are con~rertsd into al~ ~d ~:ombine~ ~ith the ~bo~r~ ~loted Terminal Doppler Weather R~dar ~icrobur~t ~h~pe~ ~ Iow Level Win~ 8hear Al~rt y:~t0m microbur~t ~h~pe~ and output as a combi~ation of a1an~l3. ;~
~l) Generation of ~u~way ~pecific Alert~:
~a) ~ fi~:d ale~t~ t~lt would be ge~e~ated ~:
individually ~ ~DWR aJ,a ~alidate~ I.LW~8 Dlicrobur~t .: ~
.
:: ~ : shapes. ~rhis i~ ~0~5~ by the i~here~t !rDWR logi~ srhich ~;
S ~ ri~ th~ i~tor3ec:tio~ o~ a give~ ~pe ~th a~ ~ale:ct -;;
box" tninally ~ ~eat~gle ~rousa the oper~tional w~y path ~ utical ~ile to ~ither ~ide z~nd "r, i~.
xtondi~g to 3 ~I.Mi off tho runway ~na). This i~ ~o~e or e~ch ~icro~ux~t ~h~ps. Eq!h~ LLWA~-ge~lerated 20 runYay-orie~t~d-1O~ (ROL) ~alue~) ar~ only U8e~ ~rhen L~ Dlicrobur~t ~hape i8 g~a~tod - but th~ ~ot -validat~d via~the syD~atry test algor~th~ hes the ;-;~
o~rerall alert Sor t;)~9 give~ opexz~tio~al ru~way i~
computed by fi~ding the ~'wor~t-case~ magnitu~e snd ~s "fir~t-e~counter" locatio~: from all the "i~lter~sti~g~
~hapa~ and the ~L ' 8 f or the ru~way. -~

::

2 1 2 4 ~ 9 3 ,~ ~?

~2) Displ~y Infor~ation:
(~) Ths bovQ logic is for gener~t~ng the runw~y alerts Th~t information is then relaye~ to the r~bbon disp1 y t-rminals for the ~ir traffic S ¢ontroller~, ~ho~ then transmit it to any impactod aircraft Th- ~me information is a}so displaye~ on the~g-ographioa1 ~ituation displ~y by "light~g-up"
; the~appropri~t- runway~locations b) Th-~TDWR~and vali~ated LLWA8 microburst 0~ shapes~re~a1so~disp1ayed on the geogr~phic ~isplay ter~inals The ~ tioned "worst-case^' m~gnitude ana "rir-t--noo~ r"~10gio is rurther appli-d ~own-stre~m ~fter th- gust-front~integr-tion alerts are separate1y l5~ qenerated ~ That ;~is,; there can - and often is -muitlpl-~t~p0~ of ~lerts for given oper~t~oD~l r~n q ~ ~ d n,~to; ~Yoid user-interpretatio~ ~a~
co~fusion issues,~only one ~lert;i~ generated for a qi- ~oper~tio~ ruDw~y ~t a~gi~en~time Tb~r-fore, 20~ th-~ bov- logic~ i8 appli-d ~for~ lerts for a Iq~t i~ 1-rts ~r-~-ep~r~t-ly ge~erat-d for ; losses ~i¢robursts~otc ~n~ g~in~ (gust fronts, et¢ ) t~en ~ singl-~"worst-¢~se" ~lert~is gen-r~te~
~ow-~-r,~ mi¢roburst~a1-rts~10s~ 30 k~*ts) alw~ys ~ take precedeDoe~ Tbat iB~: if there is concurrently a 35 knot 108~ ~nd~45~not gain - the 35 knot 109~ is , , ~

.

WO94/08312 2 1 2 ~ ~ ~ 3 PCT/US93/09128 ~ . .~

used. Thi~ is bec~u~e a ~ind shear that ~oula generatQ a very h~z~rdous lo~s ~i.e. ~ 30 knots) i~
~onsidered to ~e more ~ignificant ~or the aircraftO

Ad~itio~l~Data A~oui~it~on 8ubnYstem~
S ~he above deacription of the impro~ed lo~-level : ~: wi~d ~he~r a~ert ~ystQ~ 100 i~ ~imply e~emplary of the typo of ~iatio~ weather ~pparatus that ~re ~ailable for u~e i~ i~plem~ti~g the ~irt~l re~lity im~ging y~t~m 10 i~ ~ aviation weather applicatio~.
: 10 Ad~itional dat~ ~cquisition appar~tus c~n inclu~e lightning det-ctor~ gust fro~t tra¢~i~g ~y~tems, w~ther radar:~to~identify the prose~ce and loou~ of storm~ cell~ ~n`d~:~ precipitatio~, icing con~ition det-ction systems, niror~ t tr~c~ing ra~r, etc. B~ch . of th-~e syste~s produ¢- data i~ic~ti~e of the pr~en¢e, ~lo~cu~ ~atùrs ~nd ~everity of ~ariou~
~eteorological~ phe~iom0na: of i~terest to a~iation op~ration~ D ~dditio~, ~opolog~cal d~t~ 3uGh a~ a LANDBA~ imaga~ of; the 1and~ ur~ace~ ~within the 20~ pre~etermi~e~ ~ mult1~i~ensio~ paoa i~ al80 aiI~ble. ~;Other: ~p~tial ~ features of the multidimen~io~al ~pa~e, sucbi as aircraft oparatio~, :: ~ restrictsd air~p~ce, airport locatio~, eito. ~re ~lso d~it~ inputs th~t~re a~ailabl~ i~ the for~ of ~tatic or dyn~ic dat- ~ro~ exi~ting i~strua~t~tioD or iDpUt ,.

WOg4/08312 PCT/US93/09128 212~ 3 to graphics subsyJtem 2 a8 initi~lizatio~ data In ~um~ary, there are nu~ereu~ sourceY of the data rel~ant to the user'~ need~ and the graphics ,.
sub~ystem 2 integr~t-J the~e data ~ources a~d filter~
the received data to create a simplifiea image of the multi~imensional ~pace for the user to en~ble the user to perform a desirea tas~ ~ithout ~ei~g overwhelme~ by the quantity of data or ~ithout h~ving to ignore major ~ources of ~ata due~to the u~er'~ bi}ity to absorb ~nd process;the~ ~ntity of data that i9 ~vail~ble Gr~D~i¢~8ub~y~t _~Architectur Figure l also~;~illustrate~ ad~itional detail of ~n implementation~of~;tb-~ graphics subsy~tem 2 ~he virtual~ r-ality~ ~imaging system 10 can serv- a 15~ plurality of~ u ers, ~ith each user def~i~g a particul~r~ imag-;~ -t ~th~t i~ to be ~isplay~
Ther-for-~, the graphics subsystem 2 ¢~ ~be equipped with~ a~ ~b-r~of~;graphics proces~i~g apparatus 31-3m Each of graphics ~proc-ssinq ~ppar~tus 31-3D~ xecei~r-s 20~ dat~ i~put from~on- or more data acquisition apparatu~
21-2n ~ in the~ form of r~r data or c~mpact data r-pr-J-ntations ~ Sh- gr~phics pro¢es9ing apparatus 31-3m con~rert~ ~ the r-ceived data into image~ a~
described below llithin ~ ~ graphics proc~s~ing app~ratus 31, . .

, , .
: :

W094/OX312 PCT/US93/0~128 21245~3 graphical object generator mo~ule 4 runctions to co~vert the r~w data or compact ~ata representation~
received from an aq~ociated data acquisition apparatuB
21 into graphical ob~jects that are later manipulated S to produce the required images Each graphi¢al object ; generator ~odul- ~ includes a plurality of graphic~
obj-ct gen-rators ~ that are ~escribed ~n addition~l~detai~l~b-lo~ The gr~phical object~ th~t ar- produc-d by~ graphical object ge~er~tor mo~ulo ~
10~ ar-~tor-d~in~datab~e 3, along ~ith viewing data input by u r;i~terface 5 The u~er input i~terface c~n b a~imply~t-rminal ~evi¢e, ~uch a~ a ~eyboard, to ~d-~ine ~ ~i~gl-~us-r selected view, or can be a evi¢-~to input~a~continuou~ ~tre~ o~ ~ata in~icative lS~ ~; o~ ¢ontinuou ly;~¢h nging us-r defi~ed ~iew Thi~
latt-r ~evi¢- c~n~b-~ s-t of sensor~ ~orn by the u~er that ~-D~- the~usër;'~ h-ad po~ition to thereby e~able the virtu~l re~lity;im~ging sy~tem 10 to prese~t the in~tant~neous virtual fioI~ of vie~ to the u~er th~t 20~ pres-ntly~ th- us-r'a fiel~ of vision Th-~ atab- 3 i~ ~lso conne¢te~ to pr-sentation subsy-tam 6 th-t conve~rts the ~ta ~tored in dat~base 3 ~into a vi8u~1 image for the u~er ; Pr-~-ntatio~s~csy~t _ 6~includ-- lement 301 ~hich 25~ ~functions to initialize the visual ~i~play that i~ to , .
~ be produ~ed for the ono or more users Thi~ functio~ ~;

`2~ ~3~3 -~6- '~

cle~r~ the proce~ing ele~ents t~t co~pri~e the gr~phic~ proces~i~g ~pp ratus 31 ~nd determine ~hich of the set~ of ~ta ~ets cont~ine~ in ~atab~e 3 ~re to be u~e~ to pro~uc- the gr~phic im~ges for the selected user Blement 302 determiAes which .

charact~ristics or par~meters of the data cont~ine~ in the d~t~b~e~3~ar-~to be displ~ye~ to this p~rticul~r de~ign~t-d u~-r ~Th-~d-termined p~r~meter~ ~re then tran~ported ~long with the r~w dat~ obtained from 10~ ~ d~tab~- 3 to~the rend-r gr~phical ob~ect3 element 303 which perform~ tbs ~d~t~ merqing, tr~nspo~ition ~nd whate~-r~other ~proc-~in~ steps are required to produ¢e~the~isu~l ~im~ge The im~ge th~t i~ produced by~th-~ren~-r~graphic~l object~ -leme~t 303 i~ then 15 ~ ~transmitt-d by~th-~appropri~t- tran-mission media to the~displ~y ll~th~t corre~ponds to the p~rticul~r w ~`Pre~-ntation ;~
EYOmP10~ of~ th- views that ~re cre~te~ by this 20~ ppar~tu~ re ilIu9tr~t-d in Figures 13 - 17 Th-~-~iews are in the conts~t of ~n~irport ~e~ther s~tQm, .
~h-r-in th-~ispl~y~ illustr~te ~ over~iew ef the ~eather in a pr-d-termined sp~ce, ~ viewed from ~bo~e in Figure 13~an~-in ucoe~-ive~view~ of Figures 1~-17 that ~r- pre~-nt-d to ~ pilot or ~n ~ir traffic ::
: :: ~

~:: : :::

: :: :

W094/08312 212 ~ ~ 9 3 PCT/US93~09128 _57_ controller illustr~ting the flight path ta~en by an aircraft to approach an~ line up ~ith a particular selected runway 79 ~t the airport As c~n be seen .. .
from th~se views, there are a plurality of ~eather S phenomena in the multi~i~en~io~al space The ~e~ther .
phe~ome~a i~clu~e win~ she~r evonts 91-98, thunaerstorms P ~n~ gust fronts G The ~isplsy illustr~tR~ not ~the ~ph~enome~a per ~e but filtere~
versions th-reof~ that indicate to the pilot of the ~ lo aircraft~ only~the significant features thereof in ;~ order to enable the pilot to avoid ~ny section~ of this phenomena that ~re ~ost d~ngerous to the op-rat~on of the ~ir¢r~ft In p~rticular, the thund r~tor~s~P may~include ~in~ shear eve~ts 91-98 5~ that ~re o~tr~mely~nqerous for aircraft oper~tions~
h-~view from~th- cockpit of the ~s~ther phenomen~ m~y be~totaIly ob~cured~due to r-in, fog or cIoua~ or snow na the illustr~tions provi~e~ in Figures 13 - 17 ~re in~icative of how~th-Jo~ uaIly obscuring e~ents c~n 20~ 1imi~ated by~the~apearatus of the virtual reality L~ag1ng -y-te~ to provia~ the pilot with a clear ln~ication o~ tbe~e~iste~ce of hazar~8 i~ the path of `: :
:;
the aircraft or ~djacent thereto The pilot can therefore avoid these haz~r~s u~ing the virtual ~25~ re~lity pr-oente~by the appar~tu~ of the pre~ent inventio~ By fly~i~g along th- clear flight p~th as ':

~ .
~:
.~ ' ~o 94/Og31~ PCI/US93/09128 a93 -sa- ~

indicated by the ~ play, the pilot can avoi~ ~11 we~th~r ph~no~e~la that are ob~cured ~y t~e visu~lly ocoluaing p~e~ome~a without havi~g to b~ i~structea by the ~ir traffio co~ltrollers. Furthen~or~, t~a ~ir S traffi~: oo~trollers ~ m~e U8e of the oapability of this ~y~t~ to vi~ually deten~ine propo~ed, flight p~th through the ~r at~er to i~antify preferr~d route~ for aircr~ft operation~. This c~p~ility ca~ 3~e initi~t~d via user i~terf~ce 5, ~h~rein a~ ~ir traffic 10 ~ c:ontrelleF mo~e~ ~ cursor OD~ t~e ~cl-een of a di~pla~
uch ~s 11, to ~lect one of the plur~lity of ~irc:raft ~ :'.,:.
A in the ~ult t di~en~io~ pace . q!hi~ ~lircr~ft elec~tion i~ ~tr~slate~, u~ g aircrzl~t po~ition, ,ltitude ~d~ he~ g data received froDl ~ aircraft 15 tr~clci~g ra~ar aæt~ acgui-~ition ~1~8y8tem ~d ~tored in ~at~ 3, ~i~to ~ ~et of coordinate~ i~dicative of a poiIlt i:~l th~ ~ultddimensional ~pa~:e. A pre~efine~
:fiold~of view ~for an aircraft o~ A parti~:ul~r typ~ i3 ~also r~tri~re~ fro~ the ~t~ba~Q 3 ~ use~ ~o ¢re~t~
20 ~o gr~p~lc ~ge ~or th~ us~er.
: ~ ~apl- of a ~r~ sually ob~Gu~ing ~e~t i8 precipitation. The ra~ge of in1;~ ty of prooipitatio~ ¢~ ~e ~ to a plural~ ty of c:ate~orieJ, or e~Zlmple on ~ range of 0 ~:lear1 to 6 25 ~na~ty). A lev-l ~ pr~cipitatio~ a3 characteri~d by cld and~or light rain that cau~e~ ~om~ ~î~imum WO94/08312 2 1 2 4 S 9 3 PCT~U~93/09128 _59-impact on visibility, such that aircraft flying t~rough 1~Ye1 1 pr~cip~t~tio~ usually ~ill roly oa i~struments for guidance rather t~t e~clu~ively on ~i~ual guid~nce. L~el 3 preoipitation is ch~racteri~a by cloud~ ana ~oder~ts rain ~ith a ~ore ~ignific~nt i~pact o~ ~i ibility. ~ight~ing i3 possible in lQ~el 3 precipitation a~d often e~nate~
form the higher l~vel precipitation regio~ that are ~: typ~cally embet~ed in a le~el 3 region ~n~ can ~tri~e . ~.
lo out~ide the bigber level region. Aircraft can u~ually fly through level 3 precipitation but it is typically : . .
.:a~oi~ed ~he~ev-r~po~sible due to the air turbule~ce encount~red therei~. A level 5 pre¢ipit~tio~ region ch~racteri~e~ by~clou~s, hsavy ra~n, ~n~/or hail ~ith;1ig~tni~g ~a~h-a~y turbule~c~ ofte~ pre~ent. A
- : ~
~: ~ : le~el 5 regio~of precipitation repr0~e~t~ n region to a~oid~due to the hazardq encounterea in flying through this region. ~ :
: In visu~ll~ repre~e~ting these ~riou~ regio~s of precipit-tion,~the level l ise-~urfa~e represe~ts the : : rough ~estent of~the ~eather", ~hile the high~r level region~ r-pre~t "~eather imp~cts~ air~pa~e~t~t lie : ; withiD the 1eve1~1 regio~ ~e i~o-surf~ces that are ~-~
isplaye~ on ~isplay ll ~a~ be op~gue or ~ami-2s transparent. If opague, only the lowe~t level ~: precipitation i-o-~urface i8 disp1~yed since the ot~er ,. ,:

WO94/08312 2~ 3 PCT/US93/09128 higher level regio~3 are ~e~tea i~qi~e of this ~o-surfa¢e ~nd oa~not be ~se~. If a s~ tran~p~rent di~play i~ ~electa~, then the neste~ regio~ of higher precipit~tion ¢a~ be ~ee~ through the semi-tran~parent . ~:
S ~xterior i~o-Jurfnoe a~ dar~er i~osurface~ or regions di~playea by ~a iso-~urface of a oontrn3ting color. .:

~enderinq Prooess ~n~ A~p~ratus ~ ~he pre f erred embodime~t of the virtual reality : ~ imaging qystem of thQ present in~ention makes use of weather phenome~a a~ one of the objects di~played to : the user. In oraer to better understan~ the ~peration of the gr~phics subsy3tem 2, the re~Bri~g of a ~icroburst ~h~pe is de~cribe~ ~n a~tional ~etail.
Th~ co~oept of~r-ndering ~ rel~t~d to app~ratus that 15: ~ ¢reate~ ~ syntbeti¢ image of a~ ob~ect. The renderer ppar~tu~ creat~J ~ sh~e~ synthetic i~age of the object based upo~ threQ-dim~n~io~al geometric . d- criptio~ a~d~finition of ~urface attribut~s of , thQ ohj-ct a~d~a mo~01 of th~ illu~i~a~ion pre~e~t in 2~0 ~tA~ spaee i~ ~hioh th~ object re~ e fin~l imag~ produe-d by the render~r appar~tus i~ spatially correct in that;surfaces are ord~r*d corr~ctly from the ob~erver, ~a tho surfa~es app~ar illumi~ate~

withi~ the ~cope of the illuminatio~ model. urf~ce~
2~ may be ai~pl~yed in ~ man~er to enhance the t~sture ef W094/0831~ 2~2~ PCT/USg3/09128 the ~urfa¢e to highlight that feature, or reflection images on the object c~n be displ~ye~ Renderer 303 . ~
c~n be ~ ~ep~r~te proce~ing element or c~n be software running~ on a proces~or shared by other S elements ~ispl~y-d i~ Figure l If ~n object is to be , rendered, ~ de~or~ption of the object i~ p~sed from dat~ba~e 3 to ~th- r-n~erer 303, where the object efinition is~-rge~ith other so retrie~e~ object d-finitions to~cre~te the im~ge ; 10 ~ Tbi~app~ratu~ function~ in ~ m~nner ~alogou~ to :
the operation of a camera The camera'~ position, ~iadng direc~on~nd type of len~ must ~ll be specifi-d ~to~produce tho ~i~ual im~ge Thi~
informatio~ i~ cre~t-d by d-termine ~iewing p~r meter~
15- ~-lem-nt~302~ fxom-the dat~ ~tore~ in d~t~b~e 3 by the u~er vi~ u~er~interf~c- 5 The ~io~ing p~r~meter~ ~re used by~the rHnder-r 303 to deter~ine the field of view~;~of ~th~ cam-r~ and to delete object3 or section~
ot~ob~oct~t~at~nr- ob~cur-~ by the~pr-sence of other 20 ~ objects~th~t~ r~ located closer to the camera and in the~lin-;ot~ ight of the o~mern~ As the camera move~
along a path, the c~mer~ ~ie~s~ the ob~ect~ in a different per~peotive The renderer 303 create~ an imaqe for each ~predefined~ iDtor~a1 of time and/or space a~ the oamera traverses the;path The repres-ntation of ach ~ob;~ect in the - .:

:.

21~93 -~2-predetermi~ea ~ultidimensio~al 8p~0e ig accomp~i~he~
by defi~ing the o~ject ~ an intercoDnectio~ of ~
plurality of polygon~ an~ line~. For ~ polygon that compriss~ a tri~ngle, it~ definition i8 acc~mpli~he~
by specifyi~g the lo¢atio~ i~ three-di~en~io~al ~pa~e of the three vertices of the triangle. ~he renderer 303 uses the ~ata ~efining the three vertices to determi~e ~hether the two-dimen~ional space ~n¢ompassed by th- three Yides of the triang}e that : 10i~terco~ect the three vertice~ i9 within the fiel~ of view of the c~mer~. If 90, the r~erer ~u~t ~190 det-r~ine ~heth~r tbe tria~gle i~ parti~lly or fully obscured ~y other tria~gle~ alreaay retrievea fro~
: the ~at~base, ~hich tri~ngles define the ~urface~ of 15~ other o~j~cts. Aaaition~l co~plexity i~ ~dde~ to thi~
ta~ by the i~clu~ion of color, te~ture, opacity a~
defini~g terms to~the object. When the renderer 303 traver~e~ ~11 o~ject~ d~fined i~ the dat~ba~e ~nd ~ i~cludea within the field ef vision, the graphi~al ; ~ ~ 20~ ;image i~ co~pleted and th~ r~ ultant i~age i8 ~::
tr~nsferr~d tQ the ~i8play A ~iorobur~t graphical im~ge compris~s ~ :~
coll~ction of t~e prl~iti~e graphic~l object , such tri~ngl~ ~nd lineq. If tri~gle~ kre u~ed to define ~25~ the surface~ o ~a microbur~t, ~ plurality of tria~gles are a~embled together to creats a multi-f~ceted : '' 21~3 serie~ of surfaces to project the image of a ~olid object D~ta ~efining the ~ertices of each tri_nglo and other relevant surface features, ~re ~tore~ in the gr_p~ical obj-ct s-gmont of the d~t_ba~e 3 ~hen the ren~erer 303 tr~versos the aat~base 3, a user -~
.
interface virtual r-ality ~efinition ~at_ba~e i9 queried to identify the filter p~r~meters th~t ~re us-~ to define;th- ob~-cts _nd features that are of interest tc the speciflc user These filter 10~ ~ ~p~ram-ter~ ean b-~microbur~t agnitude, pro~imity to an ~irport run~y~or ~:ircr~ft flight path, direction :
of move~ent~,~otc The a~ditional object~ _re _180 r triev-d from the~d tab--- 3, such as other airer_ft, ; precipitation, qust fronts, terrain features, etc 15~ B~ch obj-ct is~efin-~ in terms of the po}ygonal hapes ~nd~; th-ir~ location ~nd oxtent ~thin tho pr-d-fined~ volum~ Additional primitive~ can be includ-d in~this sy~tem,~uch as object tran~parency, n~tiv-~color~,~ gr~phic~l r-pre~entation color, et¢ -~

20~ Co~p~ct~D~t~Re~r entation Conv-r~ion To Object Model ~In ord-r ~;~to illustr~te the u~e o polygonal h~pes to d-fi~ object surface~, the microbur~t sh~p-s de~cribed~; bov- ~re illu~trated in three-imen~ional ro~m;~ Sh- compact ~ata repre~entation of 25 ~ the~microbur~t ~ho~n in Figur- 8 i~ a band-ai~ ~haped -:

.~

~ ~ .
: :~

WO 94/Q8312 PCl/US93/09128 2 1 2 6 4 ~:?~

t~o-dimen~io~al regio~ on a urfa ::e that i~ u~ea to i~dic~te the lo¢~tion ~d ~xtent of ~ ~iation hazard. Thi~ shape c~n be ~efine~ i~ term~ of two poiat ~d a r~aius as not~ by thei r Carte~
600rdinat;e~l - (2el~ Ye~ d ~e2~ Ye2) ~ a~d ~ r~
The Dlag~itu~ of the wi~la ~hoar i~ thi~ ~icrobur~t i~
~a~aitio~al ~lat~ that defi~sa thi~ shear eve:~t ~n~
. .
may ~ot be used initi~lly t:o produce ~ repre3entation of the wi~d ~ear eve~lt, ~i~ce a~ de~cribed ~bove, all wind ~bear ~rent~ 2CCQ3S of a predeter~i~e~
Dlag~itude ~r~ c:on~ideraa to be a hazar~ houl~ b~
~ i~pl~ye~. }Iowe~er, the usar's filter lle~iaitio~ may :
i~clud~ ~aicrGburJt~ ~everity para~eter~ that n~o~ulat~
tk~ ~t~n~ard ~ d:: shear e~e~t thr~ho~d ~t~ to 15 aispI~y ~oro or e~r~r ~:Lsrobur~t~ 0 as a fuDctio:ll of :: thQ u~er~s filtQr. Th~ Cartesias Coor~i~ate y~teD~
: ~
finea 111 rel--tion to the ~ultidim~ional ~p~ce.
A microburst. ~ i~ a ~ur~c~ rel~ted ph~o~a, in that the ~ircr~ft ~haz2r~ o~ly ~ ts whe~ the a~ raraft 2 ;0~ in ~ clo~e . pro~cimity to the grou~l 0 ~uch B8 in a final appro~c:h to ~ airport ~Ay or i~ed~ately UpOll tak~off. ~Iosre~er, ~ graphic:~l image of th~
, microbur~t must h~ three-di~en~io~al ~te~t to ~Qure t~z~t the u~er ca~ vi~ually~et~ct it~ pr~ e~ce 25 in tho di~pla~ 11, esp~cially ~rhe~ ~iewed fro~
ob~u~e zl~glQ. Th~ ~ompact data repres~rltatio~ of a :

WOg4~08312 PCT/US93/09128 ~ 212~5~
. -S5-~icrobur~t a~ 3hown in Figure 7 provid~ two-dime~ional ~efin~tion of the Yurf~08 loous ~d e~te~t impacted by ths microburst. The ~urface is typically Dot featurele~ and thi~ two-dimen~ional compact data S representation ~ust be tran lata~ i~to ~ three-di~Qn~iona1 repre~e~tatio~ of the e~snt. Therefore, the a~tab~se 3 i~ queried to obtnin topologic~l a~t~
that def ines the ~urface feature in a~d arou~d the microbur~t imp~ct ~rea. The renderer 303 map~ the two-di~n~ionaI Iocu~ ~d extent dat~ i~to a three-.
dimen~io~al repre~entation of the actual ~urface area : impact~d ~y tbe;microburst ~ illu~tr~ted in F~gure 8.
.
: In ad~ition, thexe ~a~ b~ a ~pre~i~g effect ~t the ~-surf~oe ~nd the ~icro~ur~t may ~a illu~trated i~ a ~
:
: 15 ~anner that the top of the microburst is ~lightly ::
: ~ ~
~; mall~r th~ it~ ~xt~nt ~t the ~urfaoe, ~omewhat like : a plato~u.
; : Figure 7-12 illu~tr~te detail~ of the method use~ by this apparatu~ to coDvert t~e two-~im~nYion~l : 20 comp~ot d~ta repre~tatio~ of a ~icrobur t into a ; : ~:three-dimen~ional~gr~phic i~age represQ~tatio~. In Figure 7, ths~ two-~i~en~ional ~aadai~ ahape i~
r~prese~te~, includi~g ~ ~atum poi~t~ Pi ~nd ~urf ace ~ormals rQpressnt~tive of the two-dimQnsional ta~g~nt . .
: 25 vector at each~datum point on the curve~ Once eaah datu~ poi~t i~ identifi~d, its loc~tio~ o~ the three~
~ -, '~-`::

2~24~93 -66-dime~io~al ~urf ~ce i~ defi~ea such th~t Pi = (xi, yi gi). Th~ seri~Y of datu~ poi~ts ~ then be illustrated diagrammatically i~ Figure 8 ~s a perimeter li~e that follows t~e topology of the ~urfaoe to i~clude th~ locu~ and e~t~nt defi~e~ by th~
t~o-~imen~io~al ban~aia ~hape. The series of ~atum poi~ts 80 defi~aa in thr~e-~imen~ion~l ~p~e ~ach hnve the property that two aata component~ (~i, yi) de~i~e t~e datum poi~t locatio~ on the bandaid peri~eter in two-dime~siona1 space. A th~rd componQ~t defines the height of the d~tu~ point above ~ ba~e level i~ the multidime~ion~ pace. ~he normal ~ctor~ generated :for each qatwm point define the "outwar~'~ airection of the ~urf~c~ of the ~icrobur~t at that ~atu~ point.
15; ~he ~orm~l ~ctorJ c~n be usad ~or sh~di~g ~na to : pro~ia~ th~ appearanGe of 2 ~mooth surfa~e.
once the s0rie~ Pi o~ ~tum poi~t~ ~re ~efl~e~ ~8 dQscrib~d abo~Ye, a ~e~o~d serie~ of ~atu~ poi~t~
: : ~re defi~d in thre~-dime~sional spac~ to defi~e the ~ . .
20: top peri~eter o~ th~ microburst. Figur~ 9 illustr~t~ ~
:
th~ rel~tio~ship~ of the rir~t aa~ eco~ ~er~e~ of datu~ point3 ~ ~two-di~e~ional :3pa~a. ~he se~o~a ~exie~ of ~atum ~point~ ~re s~l~cte~ ~o that th~ir perimeter ~efi~e~ ~ ba~dai~ shape of locu~ e~t~at 1~ than t~t o~ the perimetes defi~ea by the fir~t ~eries of aatu~ poi~ts ~o that the ~econ~ V~ries are WOg4/08312 PCT/US93/09128 212~9~

inscribed ~ithin the first series of datum points.
-Tha third co~pon~nt of s~ch datum point in the ~econ~
~erie~ of dntum point~ i~ selecte~ to be a fixsa height ~bove the ~urface locatio~ of ~ corre~ponding s datum poi~t in the fir~t ~erie~ of ~atum point~. The result~nt three-di~-~io~l ~hape i~ illu~trated in Figure ~o and:re~emble~ a pl~te~u whose top surf~cQ
follow~ the contour~ of the surface on whiah it re~t~
~ To ~able th~ renderer to p~rform its ta~k. the ;~ ~ lO microbur~t ~shapo d-fined by the two ~erie~ of d~tu~
points are co~verted to a plurality of ~urface : : -~ defining polygon~. Figure 11 illu~trat~ the :
d~finitio~ of-o~e sid~ ~urfac~ of the microbur~t while Figure l2~illu~trates the de~inition of the top of the 5~ microbur t In p~rtiGular~ ia Figure ll, a tri~ngul~r s-gment of surf~c~l results fro~ co~ne¢ting tho t~o rie~ of datum points ~ith lines t~at defi~e the edges of triangle~ A fir~t tri~gle i~ defined by ¢onnecting poi~ts~Pl, Ql, Q2) uhile ~ ~co~d tri~gle 20 ~ is defines by ~on~ct~g points~ (Pl, P2, Q2) Thi~
proce i conti~ued u~til t~e 1~t defî~ea tri~ngle coD~ect to the origi~al or ~tarti~g point3 Pl, Ql~
A ~imilar proces~ i~ us-d t~ defi~e the top o~ the microburst ~hape 25 The e~terior surf~cs~ of the microburst are thu~
~efine~ by a series of tria~gle3, each of ~hich i~
~ , ' WO 94/~83~2 PCI/US93/~912~
212 4~93 -68~

preci~ely loc~ted within t~e multi~imen3io~ pæce by the C~rte~i~ Coordin~te~ of th~ verti¢e~ of e~ch tria~gla. The render~r can then u~e the vurtice~ and the ~urface nor~;ll ~ectors to denote surf~ce~ in the fiQld of vi~ion of the user ~nd to x~pre~ent the ob~curation of one o~ject by another. O~c~ th~ fi~ld of ~ion is thu~ defiIlQa, the addit~ on~l ~ttribute~
of the Yariou~ objeot~ nre u~e~ in coaljunction with the u~er-define~ filter~ to tra~l~form the visual i~age : ~ : lO into t~:e ~irtual rea1ity def in~d by t he u~er . In particul~r, the types of object~ played c~ be defined ~y t he u~r to eli~in~te the ~isually -; ~ ob~curi~g effe~t~ of precipitationD or fog, or clouas~In ~itio~, th~ 3urfac~ defin~d by th~ tri~ngle~ a~a th~ surf~ce ~orm~1 veotoE-q c~ be ~i~u~lly di~played ~: u~ing~ S~al~g: a8 ~ fu~¢tio~ of t~ ~agnitude aI~
: directio~ of each 2~urfaae ~efini~g vector to proYi~e ~: tb~: u~er with~ accur~t~ thre~ nsis~ oli~
x~prQse~tation ~of the objec:t. Furt3~rmore~ tiDIe 2:0 ~qu~tial~ v~lues~ for the location of eao~ ~at~m point .
e u~ to prov~e ~ ~o~i~g image~ repE0~e~tati of the obj~et to illu~trate its ~ov~ment iD th~ :
~ultidi~n~ional ~ spac~ over ~ r of time-:~egue~tial ti~e inter~al~. Thi~ ti~ 3erie~ of ~ata ~ -2 S poi~t~ ca~ al80 ~ e~trapolated to prediat the future ~' movemellt of th- luicroburst.

, .

wo 94f08312 2 1 2 4 5 9 3 PCT/US93/~28 ,~ .~ ,~

Display Ex~E~e ~ igures 13-17 ~llustrate ~n e~mp}e of a ~erie3 of displays thnt cæn be usea in ~n aircraft ~eather display application of the ~irtual reality im~gi~g ~y~tem lO~ Figur~ 13 illu~trate~ a top view of a pr~d~t~rm~n~d ~paco, ~hich ~iew i~clude~ a plurality .
of feature~ ~nd p~sno~ena. The predetsxmi~ed spac~ i~

deli~eat~d by bound~ry li~e~ B a~d range circle~ R1-R3 : : : c~n be include~: to ae~ote range for~ a pr~determi~ea : .

point locat~ i~ the predeter~i~ed space. Temporally :
con~tant faature~ looat~d in this spaoe are road~
:i H6~nd natural topological features, such a~ mountai~ ~'.7'"
M. Otber f0atur-~ ~how~ on Figure 13 ~re airport -~
run~ay:79 un~ ;~ircr~ft A. T~por~lly a~a ~p~ti~lly ~' 15 : varying pheno-~n~ that ~ra pre t i~ thi~ ~pac~ are ; .
regions of precipit~tio~ P, ~ind sh~nr ~e~t~ 91-98, :~
~nd~ gu~t front~G.~ Collecti~ely, tha~e ~lement~
represe~t~the item~of i~tere~t to a rcraft operatio~s ~ ~ , ~ in the pre~etor~ined ~p~c~. It i~ ob~ious that a `:;
, 20: : level o~ precipitatio~ eleat~a for th~ op~r~tio~
of ai~pl~y ll~ to:~:deli~e~te in Figure 13 o~ly t~è

e~te~t of ~the~ seIected le~el of preaipit~tio~ to ; mini~ize the comple~ity of the di~pl~y 3ho~n i~ Fi~ure 13. This feature~e~able~ tbe ~iewer to ~i~imize the 25; oxtent of the preoipitation region~ d~pI~ya~ to o~ly the levels of ~tor~t. For e~mpl~, if o~ly le~el 3 . . t, 2124~93 or higher rogions of precipitation ~re ~ispl~yed, then the visu~l obscur~tion pre9ente~ by the l~vel 1 ~n~ 2 precipitation ~re ~-lete~ from the ~i~play thereby enh~ncing the ~visibility for the ~iewer, 9uch ~9 a pilot A more useful ~i8pIay for ~ pilot is 8hown in Figure 1~ ~here the~u9e~ via user interf~oe 5 ~efines .:
~ poi~t in th~ predet-rmine~ sp~ce ~nd ~ fiold of - ~ .
~iew ;Thi8 d~t~ is tr~nsl~tea into ~ perspecti~e S0 three-dimensional type of display to illustrate the view from the~selected point, with the ob~ouring ~ata filt-r ~ out by~the~r~nd-rer 303 As note~ above, the low~lev~ pr-cipit~tion ~nd oth-r aircr~ft A o~n be removo~from~the~view ~n~ only objects of intere~t ;15; ~ispl~ye~thereon~ The display thereby presents a~
g ~of ~th- potenti~lly thr-atening ~eatber phenomena that~eonfront~the~pilot, in the form of precipitation P~of~at~l-~st ~lev-l 3, ~ind ~he~r ev~ents 91-98 an~
urf~c-~topology~8; ~ote th~t~the bou~d~rie~ B of the 20~ pr-d-t-r in~d ~p-c-~ -r- ~80;8hovn to indicate the e~tent~of the~data pre8ent in th~ fiel~ of ~iew A8 the~ircr~ft a-cre~ses its ~ltitude an~ppro~ches ths run~y 79~, ~th-; field of view ~na point in t~e pr-d-termined~;8p-ce change ~s the aircr~ft traverseq the flight~p~th The virtual re~lity im~ging ~ystem 10 periodically~samples the data to update the point : : :

:

.
.

W0 94/~8312 2 1 2 4 5 9 ~ PCr/U593/Ogl28 --7 l--i~ space, fiela 9f ~iew as ~ell as ~elect~d characteri~tic:s for ~i~play to create ever c:hzulgi~g ge~ u~:h ~ that sho~m ~ Figure lS, ~9 the aircraft tra ~ar~as the predet~rmi2led space to Gir 5 the ru~nway 79;~ li~a~up for ~n approach~ The vi~w o~
~pproach i~ ~ho~m i~ Figur~ lC ~here the ruslway 79 i~
clearly s~en ~ are rogio~ of pr~cipit~tio~ P a~
wi~d ~hear eVeDt~ 96098. q!hese wi~ ~h~r event 96-~8 in r~lity ~y ~lOt b~ sible to the pilot and the ~isplay ~t ~ ~; ¢omput~or ge~erate~ ren~ering provide~
tb~ pilot with inforEatio~ of w~th~r related : phenome~ t~at~oth~rwi~e i unavailabl~. The pilot . .
: csn ~t~rmin~ ~he pro~i~ity of the ~in~ ~hear ev~ts 6-98~to runY~y in:d-cidi~g ~hot~r to ¢ontinue th~
~; ~15 appro-oh to runya~ 79. In aaait~O~, r~der~r 303 c~n be~ ~Gtivate~ ~ to~ e~trapolate th~ ~i8ti~g ~ata to illu~tr~t~ lik-ly progra~io~ of ~i~ ~he~r event~
96-98~ t~at is:~::likely to o~cur duri~g the airora~
ppro~¢h~ Thi~ preai~ta~ 3ce~rio ~an be qui~kly :displ~yed o~ i3play ll to ena~le the pilot to dQtermin~ the c10~9t approach o~ ~i~d ~hsar ~Ye~t~
: ~6-~8 4uri~g the ntire l~nding:op~r~tion. Figure 17 illu~tr~ta~ ~ti-e~i~e s-quo~ti~l ~i9play followin~
that of Fi~ura~16 to indicate the di~play that would 2~ be seen further along th~ fligbt path a~ thQ runway 79 i approachQd.

212~59 3 -72 ~ummary As ~n be ~een fro~ the aboYe e~ample~, the ~irtual reality im~ging ~y~te~ displ~y~ feature~ aad phe~om~na, that can be temporzlly and/or ~pati~lly v~rying, in a maDner to filter out the ch~racteri~tics of the features ana phenomena t~at ~re not of interest : ~ to the vie~Qr. The~image pre~e~ted to'the viewer i~
; a condensation of all t~e data oollected by the plurality of dat~ acquisition ~y~tem~, ~nd ~o~e of the O data pre~e~tea~represents ~eature or phenome~a that ~re ~ot Yisible to the ~iewer ~ith the naked eye.
.,, . ~hus, this ~pparatu~operate3 in real time to provid~
ach u~r with a: oustomized vie~ of a predeter~ined ~;~
, spac- to enabl-~th~ u~or to peror~ ~ ~Qsired t~sk.
5~ ~ Whil- ~ ~pe~clic ~boaime~t of ~hi~ in~ention h~s ;~
been~di~los-d,~it i~ e~pected that those ~illed in :ths ~rt ca~ and ~ill de.ig~ alternate e~bodiments of ~:
"~,..;
:
thi~ i~vention that fall within ths cope of the ppended olai~. ~

: ~
;

...

',:

Claims (36)

I CLAIM:

1. Apparatus for presenting a user with a virtual image in real time of phenomena located in a predefined multidimensional space, comprising:
means for generating data, in real time, indicative of at least one phenomena extant in a multidimensional space, which multidimensional space has predefined extent in a plurality of dimensions;
means for storing data defining a plurality of characteristics of said phenomena that are to be displayed to a user;
means for extracting data that satisfies said plurality of characteristics, defined by said stored data, from said generated data; and means, responsive to said extracted data, for producing an image representative of a three dimensional view of at least a portion of said multidimensional space to display said phenomena, substantially temporally concurrent with the generation of said data used to produce said image.

2. The apparatus of claim 1 wherein said generating means comprises:
means for measuring present values of a plurality of parameters for at least one temporally variant phenomena extant in said multidimensional space.

3. The apparatus of claim 1 wherein said extracting means comprises:
means for storing data indicative of a predefined point in said multidimensional space;
means for storing data indicative of a field of view from said predefined point in said multidimensional space;
means, responsive to said stored predefined point and field of view data, for defining a segment of said multidimensional space representative of said field of view as taken from said predefined point; and means for identifying a portion of any of said phenomena that lie within said segment of said multidimensional space.

4. The apparatus of claim 3 wherein said generating means comprises:
means for displaying said plurality of characteristics of said portion of said phenomena included in said segment of said multidimensional space.

5. The apparatus of claim 4 wherein said extracting means further comprises:
means for defining a motion vector through said multidimensional space from said predefined point to a termination point;
means for defining a sampling interval;
means, responsive to said stored data indicative of said predefined point, said motion vector, said sampling interval and said field of view, for defining a plurality of segments of said multidimensional space representative of said field of view as taken from said predefined point at such of a plurality of successive sample intervals, as said motion vector traverses said multidimensional space from said predefined point to said termination point.

6. The apparatus of claim 5 wherein said generating means further comprises:
means for extrapolating said measured present values of said plurality of parameters for at least one sampling interval into the future from a present sampling interval.

7. The apparatus of claim 4 wherein said extracting means further comprises:
means for defining a sampling interval;

means for extrapolating said measured present values of said plurality of parameters for at least one sampling interval into the future from a present sampling interval;
means, responsive to said stored data indicative of said predefined point, said sampling interval, said measured present values of said plurality of parameters, said extrapolated parameters and said field of view, for displaying said field of view as taken from said predefined point at each of a plurality of successive sample intervals including said present sampling interval and said at least one sampling interval into the future.

8. The apparatus of claim 1 wherein said generating means comprises:
means for measuring a plurality of parameters for at least one temporally invariant phenomena extant in said multidimensional space.

9. The apparatus of claim 1 wherein said generating means comprises:
means for simulating a phenomena in said multidimensional space.

10. The apparatus of claim 3 wherein said measuring means measures a plurality of parameters for at least one temporally variant meteorological phenomena extant in said multidimensional space, said apparatus further comprising:
means for measuring a plurality of parameters for a phenomena comprising at least one spatial feature extant in said multidimensional space;
and means for storing data defining a plurality of characteristics of said spatial features that are to be displayed to a user.

11. The apparatus of claim 10 wherein said multidimensional space comprises a volume of airspace, said temporally variant phenomena measuring means comprises:
at least one weather data acquisition system for producing data indicative of weather phenomena in said multidimensional space.

12. The apparatus of claim 11 wherein said multidimensional space comprises an airspace, said spatial feature measuring means comprises:

topological data acquisition system for producing data indicative of surface topological features in said multidimensional space.

13. The apparatus of claim 12 wherein said characteristics storing means comprises:
means, responsive to user input data, for defining a set of threshold indicative of weather phenomena magnitude.

14. The apparatus of claim 13 wherein said predefined point storing means stores data indicative of a position of an aircraft, said field of view storing means stores data indicative of a field of view in the flight path of said aircraft, said displaying means comprises:
means for producing a visual display for a pilot of said aircraft of said weather phenomena exceeding said threshold that are inclusive of said field of view in said flight path of said aircraft.

15. The apparatus of claim 14 wherein said extracting means further comprises:
means for defining a motion vector through said multidimensional space from said predefined point to a termination point;

means for defining a sampling interval;
means, responsive to said stored predefined point, motion vector, sampling interval and field of view data, for defining a plurality of segments of said multidimensional space representative of said field of view as taken from said predefined point at each of a plurality of successive sample intervals, as said motion vector traverses said multidimensional space from said predefined point to said termination point.

16. The apparatus of claim 15 wherein said displaying means comprises:
means for displaying in each said successive sample intervals, said plurality of characteristics of said meteorological phenomena and said multidimensional space inclusive of each said segment of said multidimensional space.

17. The apparatus of claim 16 wherein said generating means comprises:
means for extrapolating said measured present values of said plurality of parameters for at least one interval of time into the future; and means for displaying said plurality of extrapolated characteristics of said meteorological phenomena and said multidimensional space.

18. The apparatus of claim 15 wherein said extracting means further comprises:
means for defining a sampling interval;
means for extrapolating said measured present values of said plurality of parameters for at least one sampling interval into the future from a present sampling interval;
means, responsive to said stored data indicative of said predefined point, said sampling interval, said measured present values of said plurality of parameters, said extrapolated parameters and said field of view, for displaying said field of view as taken from said predefined point at each of a plurality of successive sample intervals including said present sampling interval and said at least one sampling interval into the future.

19. A method for presenting a user with a virtual image in real time of phenomena located in a predefined multidimensional space, comprising the steps of:

generating data, in real time, indicative of at least one phenomena extant in a multidimensional space, which multidimensional space has predefined extent in a plurality of dimensions;
storing data defining a plurality of characteristics of said phenomena that are to be displayed to a user;
extracting data that satisfies said plurality of characteristics, defined by said stored data, from said generated data; and producing, in response to said extracted data, an image representative of a three dimensional view of at least a portion of said multidimensional space to display said phenomena, substantially temporally concurrent with the generation of said data used to produce said image.

20. The method of claim 19 wherein said step of generating comprises:
measuring present values of a plurality of parameters for at least one temporally variant phenomena extant in said multidimensional space.

21. The method of claim 19 wherein said step of extracting comprises:

storing data indicative of a predefined point in said multidimensional space;
storing data indicative of a field of view from said predefined point in said multidimensional space;
defining, in response to said stored predefined point and field of view data, a segment of said multidimensional space representative of said field of view as taken from said predefined point; and identifying a portion of any of said phenomena that lie within said segment of said multidimensional space.

22. The method of claim 21 wherein said step of generating comprises:
displaying said plurality of characteristics of said portion of said phenomena included in said segment of said multidimensional space.

23. The method of claim 22 wherein said step of extracting further comprises:
defining a motion vector through said multidimensional space from said predefined point to a termination point;
defining a sampling interval;

defining, in response to said stored data indicative of said predefined point, said motion vector, said sampling interval and said field of view, a plurality of segments of said multidimensional space representative of said field of view as taken from said predefined point at each of a plurality of successive sample intervals, as said motion vector traverses said multidimensional space from said predefined point to said termination point.

24. The method of claim 23 wherein said step of generating further comprises:
extrapolating said measured present values of said plurality of parameters for at least one sampling interval into the future from a present sampling interval.

25. The method of claim 22 wherein said step of extracting further comprises:
defining a sampling interval;
extrapolating said measured present values of said plurality of parameters for at least one sampling interval into the future from a present sampling interval;
displaying, in response to said stored data indicative of said predefined point, said sampling interval, said measured present values of said plurality of parameters, said extrapolated parameters and said field of view, said field of view as taken from said predefined point at each of a plurality of successive sample intervals including said present sampling interval and said at least one sampling interval into the future.

26. The method of claim 19 wherein said step of generating comprises:
measuring a plurality of parameters for at least one temporally invariant phenomena extant in said multidimensional space.

27. The method of claim 19 wherein said step of generating comprises:
simulating a phenomena in said multidimensional space.

28. The method of claim 21 wherein said step of measuring measures a plurality of parameters for at least one temporally variant meteorological phenomena extant in said multidimensional space, said method further comprising the steps of:

measuring a plurality of parameters for a phenomena comprising at least one spatial feature extant in said multidimensional space; and storing data defining a plurality of characteristics of said spatial features that are to be displayed to a user.

29. The method of claim 28 wherein said multidimensional space comprises a volume of airspace, said step of temporally variant phenomena measuring comprises producing data indicative of weather phenomena in said multidimensional space using at least one weather data acquisition system.

30. The method of claim 29 wherein said multidimensional space comprises an airspace, said step of spatial feature measuring comprises producing data indicative of surface topological features in said multidimensional space using a topological data acquisition system.

31. The method of claim 30 wherein said step of characteristics storing comprises:
defining, in response to user input data, a set of thresholds indicative of weather phenomena magnitude.

32. The method of claim 31 wherein said step of predefined point storing stores data indicative of a position of an aircraft, said field of view storing means stores data indicative of a field of view in the flight path of said aircraft, and said step of displaying comprises:
producing a visual display for a pilot of said aircraft of said weather phenomena exceeding said threshold that are inclusive of said field of view in said flight path of said aircraft.

33. The method of claim 32 wherein said step of extracting further comprises:
defining a motion vector through said multidimensional space from said predefined point to a termination point;
defining a sampling interval;
defining, in response to said stored predefined point, motion vector, sampling interval and field of view data, a plurality of segments of said multidimensional space representative of said field of view as taken from said predefined point at each of a plurality of successive sample intervals, as said motion vector traverses said multidimensional space from said predefined point to said termination point.

34. The method of claim 33 wherein said step of displaying comprises:
displaying in each said successive sample intervals, said plurality of characteristics of said meteorological phenomena and said multidimensional space inclusive of each said segment of said multidimensional space.

35. The method of claim 34 wherein said step of generating comprised:
extrapolating said measured present values of said plurality of parameters for at least one interval of time into the future; and displaying said plurality of extrapolated characteristics of said meteorological phenomena and said multidimensional space.

36. The method of claim 33 wherein said step of extracting further comprises:
defining a sampling interval;
extrapolating said measured present values of said plurality of parameters for at least one sampling interval into the future from a present sampling interval;
displaying, in response to said stored data indicative of said predefined point, said sampling interval, said measured present values of said plurality of parameters, said extrapolated parameters and said field of view, said field of view as taken from said predefined point at each of a plurality of successive sample intervals including said present sampling interval and said at least one sampling interval into the future.