WO1997021990A2

WO1997021990A2 - Methods and means for monitoring impacts in vehicles

Info

Publication number: WO1997021990A2
Application number: PCT/US1996/019521
Authority: WO
Inventors: Timothy P. Slifkin; W. Glinski Lech
Original assignee: Elexor Associates, Inc.
Priority date: 1995-11-28
Filing date: 1996-11-27
Publication date: 1997-06-19
Also published as: AU1565197A; WO1997021990A3

Abstract

Events in a vehicle are monitored by producing electrical outputs representative of events in the vehicle, comparing characteristics of one event with characteristics of other events accumulated over a given period of time and determining departures of a given extent from the other characteristics as an indication of a significant event, and sending a warning in response to the indication, particularly with the position of the vehicle determined by a global position indicator (GP1) on the vehicle. In an embodiment a microcomputer (MC1) responds to the outputs of an accelerometer system (AC1, AC2, AC3) mounted on a railroad car by comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts and determining departures of a given extent from other characteristics as a failure indication, signalling a warning in response to the indication. One of the outputs is in the vertical direction to test the vehicle suspension. Position of the vehicle is captured periodically in periods of less than two minutes and sent with the warning signal.

Description

TITLE

METHODS AND MEANS FOR MONITORING IMPACTS IN VEHICLES

RELATED APPLICATIONS

This application relates to the concurrently filed application entitled Methods and Means For

Monitoring Events in Vehicles by the same inventors as this application, to another concurrently filed application entitiled Methods and Means For Monitoring Impacts in Vehicles, and to copending application also entitled Methods and Means For Monitoring Events in Vehicles filed November 28, 1995.

FIELD OF THE INVENTION

Thiε invention relates to methods and means for monitoring various operational aspects within a vehicle, and particularly for methods and means for determining if aspects of a railroad car or other vehicle is functioning incorrectly.

BACKGROUND

Various events affect the operation of vehicles, such as trucks and railroad cars, and the safety of their cargoes. For example, vehicles and their respective cargoes, are subject to substantial shocks from sources such as rear and front impacts, damaged suspensions or wheels, and, in the case of rail road cars, out of round wheels, unbalanced wheels, and "truck hunting" (the term used when a rail car's wheels vibrate back and forth between the rails rather than traveling down it smoothly. In railroad cars, efforts are made to limit damage from impacts shocks by means of cushioning units mounted between cars. As long aε these operate properly they help protect the railroad car and its cargo from the impacts that occur when cars are coupled together or from "in- train" forces that occur when the train is being pulled along the tracks.

In order to keep vehicles operating properly, it is desirable to obtain information, either in the vehicle, or at a remote locating or both, concerning events that affect the vehicle operation.

SUMMARY OF THE INVENTION

According to an embodiment of the invention, events in a vehicle are monitored by producing electrical outputs representative of events in the vehicle, comparing characteristics of one event with characteristics of other events accumulated over a given period of time and determining departures of a given extent from the other characteristics as an indication of a significant event, and sending a warning in response to the indication. Examples of particular events sensed logitudinal impacts on coupling cushions, vertical impacts on vertical suspension components, lateral impacts (transverse to vertical and longitudinal) or combinations of these.

According to another embodiment, a position signal shows the position of the vehicle with a global position indicator, and the warning is sent with both the indication with the position signal. The system turns on the position indicator at periodic intervals.

The various features of novelty that characterize the invention are pointed out in the claims. Objects and advantages of the invention will become evident from the following detailed description when read in light of the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a schematic diagram of a rail car employing an embodiment of the invention.

Fig. 2 is a schematic diagram of a system used in Fig. 1.

Fig. 3 is a flow chart of an embodiment of the operation of a microcomputer in Figs. 1 and 2.

Fig. 4 is a graph illustrating the acceleration waveform of an impact and the determination of an event as used in the system of Figs. l and 2.

Figs. 5 is a flow chart of another embodiment of the operation of a microcomputer in Figs. 1 and 2.

Fig. 6 shows details of Fig. 5.

Fig. 7 is a graph illustrating the distribution of a running number of impacts within a multiplicity of ranges over a period of time as used in the system of Figs. 1 and 2.

Fig. 8 is a graph illustrating the distribution of a running number of events within a multiplicity of event ranges over a period of time as used in the syεte of Figs. 1 and 2.

Fig. 9 is a graph illustrating the acceleration force of a single impact over a period of time as used in the system of Figs. 1 and 2.

Fig. 10 is a graph illustrating acceleration rise times relative to the fall times over a number of impacts as used in the system of Figs. 1 and 2.

Fig. 11 is a graph illustrating acceleration rise times relative to events over a number of impacts as used in the system of Figs. 1 and 2.

Figs. 12, 13, and 14 are schematic diagrams illustrating yet other embodiments of the invention.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Figs. 1 and 2 illustrate an embodiment of the invention where the events occur in a rail car. In Fig. l, a rail car RC1 carries a solar panel SPl for collecting solar energy to supplement an internal battery or other power source in an impact monitor unit MUl. In some embodimentε, such a solar panel is not used and is unnecessary. An antenna ANI connects to a global position indicator GPl which is in the monitor unit MUl and which responds to global position satellites (GPS) to establish the position of the car RC1. The output of the global position indicator GPl also passes to the microcomputer MCI, and may receive data therefrom if needed. Couplers COl and C02 join the rail car RC1 to other cars through respective cushioning units CUI and CU2 which are fastened to the rail car RC1 itself. Suspension units SU1 and SU2 support wheels WH1 which ride on tracks TRI under the car RC1.

The impact monitoring unit MUl captures all impacts above a certain level and in all directions from all sources, such as out of round wheels (vertical shocks) , side to side sway, twisting, and longitudinal impacts. It then analyzes the impacts, and queries the internal global position indicator GPl for precise location information, transmits the results to a central location, together with the location determined by the global position indicator GPl to the antenna ANI. The latter transmits the information to a central station remote from the cars. A single message is then constructed consisting of all monitored parameters

(impact, temperature, etc.) plus the current location information. The complete message is then transmitted or sent to the central base station via the internal satellite transceiver, cellular telephone, radio frequency transceiver or other similar device.

Additionally, a summary of the message is stored internally to the monitoring system MUl for later reading via a direct-connect device such as a laptop computer or similar device. The monitor unit MUl also re-evaluates all impacts and other data each time a new one is added. The system therefore includes effects due to the specific parameters of the car RC1 itself as well as its cargo. As shown in Fig. 2, the impact monitor unit MUl contains an accelerometer set AS1 composed of accelerometers ACl, AC2, and AC3, sensitive to shocks in respective orthogonal directions. For example, accelerometer ACl is sensitive in the X horizontal longitudinal direction, accelerometer AC2 in the Y horizontal latitudinal (orthogonal to the horizontal longitudinal) direction, and accelerometer AC3 in the Z vertical direction orthogonal to the X and Y horizontal directions. Hence, shocks arriving in a direction angular to any of the X, Y, or Z directions cause the accelerometers to respond to the component of the angularly arriving shock along the respective orthogonal directions. Thus, the accelerometer, ACl, AC2, and AC3 each produce analog electrical outputs representing the components of any shock in their respective directions of sensitivity.

Amplifiers AMI, AM2, and AM3 receive the electrical analog outputs from the accelerometers ACl, AC2 and AC3 and apply them for filtering by respective filters FII, FI2, and FI3 filter. Signal conditioners SCl, SC2 and SC3 then condition the signals by adding or removing signal elements. A multiplexer MX1 time-division multiplexes the analog signals, and an analog to digital (A/D) converter AD1 converts the multiplexed analog signals to digital form before applying them to a microcomputer MCI.

The microcomputer MCI processes the information from the converter AD1 and transmits the processed data to an output device ODl. The latter passes the data to the antenna ANI. The microcomputer MCI may pass the data directly to the antenna ANI. At the same time the global position indicator GPl passes its output to the antenna ANI. The latter then transmits the processed data and the global position indicator output to a central station, for example on the premises of a railroad office. The microcomputer MCI may also operate the output device ODl in the form of a tag or flag, an LED, serial communication, or a switch to notify service personnel that a failure is possible and maintenance is needed. The microcomputer MCI may also relay relevant data via a serial port directly, or via the device ODl, to the antenna ANI to another device (e.g. handheld terminal, satellite, cellular or radio communication port, etc.) The microcomputer may also process the data from the global position indicator and pass it to the antenna ANI.

If needed, the solar panel SP1 supplies energy to a solar power supply SOI that energizes, or supplements, other energy sources such as a battery BAl for all the elements requiring energy in the monitor unit MUl.

The microcomputer MCI performs the steps shown in the flow chart of Fig. 3 to apply a warning signal to the output device ODl that displays a signal representing a failure warning or send relevant information to another communication device via a serial port. In Fig. 3, the microcomputer MCI first initializes the system and tests the battery, solar panels and other operating elements in step TST. In step SIT, the microcomputer MCI sets an initial or first threshold which any signal from the accelerometers ACl, AC2, AC3 through the converter AD1 must exceed to be deemed of sufficient amplitude to exceed noise, for example 0.1 G and then passes only such data. In step PMM it enters the monitor mode.

In the monitor mode, the microcomputer MCI first determines if any impact that exceeds the first threshold also exceeds a second threshold, one sufficiently significant to warrant transmission to the central station, and higher than the first threshold. For this purpose, while continuously responding to the multiplexed digital data from the A/D converter ADl, it captures the waveform of an impact that exceeds the first threshold in step C I. An example of such a waveform appears in Fig. 4. In step MAA it determines the "shock peak" namely the maximum G acceleration amplitude or acceleration peak of any impact or shock. In step DDN it determines the "shock duration" i.e. the duration of the acceleration above a minimum threshold. In step DDV it determines the "impact velocity" or "Δ velocity", i.e. the integrated acceleration from a time

tl to a time t2, specifically f"adt = Δv , where

a is the acceleration in G's, v velocity, and tl and t2 times during the impact, as shown by an area AR1 under impact curve ICI between times tl and t2 in Fig. 4.

In step PPM, the microcomputer MCI also turns on the global position indicator GPl for a predetermined interval sufficient to capture position data, and then sets the indicator GPl to turn on at periodic intervals, such as 30 seconds. Fig. 3A illustrates details of a step Sll after step PMM.

In step TDA, the microcomputer MCI then establishes a "impact total" or "acceleration total" or "shock total" which is a total or weighted total, of these determinations and in step CTD then compares the "impact total" with the preset "second threshold" which is substantially higher than the first threshold. If the impact total exceeds the higher second threshold, the microcomputer MCI deems the impact as a significant impact. Then, according to one embodiment shown in step OUT, the microcomputer MCI transmits the data together with the data from the global position indicator GPl to a central station via the output device ODl and the antenna ANI. In step EXC if the impact total is less than the preset second threshold, the microcomputer MCI, in step NOS, does not send the data to the central station.

According to another embodiment, the computer MCI analyzes the data to be sent, i.e. the data that exceeds the second threshold, further before transmitting it. For this purpose it utilizes both the data which exceeds the first threshold and that which exceeds the second. It stores all the data that exceeds the first, lower, threshold. It further analyzes the data by comparing all new data that exceeds the second threshold with the history of prior data which is continuously updated with all data that exceeds the first threshold. This is done as shown in Fig. 5.

In Fig. 5 the microcomputer MCI adds all new data at step SIT that exceeds the first threshold and stores a history of all new events, e.g. impacts etc., that exceed the first threshold, over a given running period such as the last thirty days. This is done in step ADD where it adds the data and in step STO stores it. In step CMP it compares each new data that exceeds the second threshold from step OUT with the historical data of the last 30 days stored in step STO. In step SGF the microcomputer MCI asks whether there is any departure that is meaningful in that it exceeds given parameters. If yes, in step ANL it analyzes the data. In step TMT it then transmits the analyzed data, together with the position from the global position indicator GPl to the central station. If no, it may retain the data, or in step TMT, it may also transmit the data to the central station for recording purposes.

An example of the process of Fig. 5 appears in Fig. 6. This example deals specifically with longitudinal impacts upon the cushions CUI of Fig. 2 to determine if they are intact and operating properly.

The cushioning units CUI are very large shock absorbers that protect the rail car RC1 and its cargo from the impacts that occur when two cars are coupled together, or from the inter-car forces that occur when a train is being pulled along the tracks. Various types of cushioning units exist. Newer ones are oil filled and pressurized with inert gaε. Older ones are in the form of large springs with some time type of damping in the form of air or oil.

The impact monitor unit MUl detects possible failure of the cushioning units CUI and CU2 resulting from causes such as leakage of pressurized gas, damage to internal valves, damage to internal springs, leakage of seals, etc. Such failure would prevent the cushioning units from protecting the car RCl and the cargo.

The impact monitor unit MUl identifieε a potentially defective cushioning unit CUI or CU2 by using accumulated acceleration characteristics of one impact with accumulated acceleration characteristicε of other impacts and determining departures of a given extent from other characteristics as a possible failure indication.

Only one of the accelerometers ACl, AC2, and AC3, namely the X-direction sensitive ACl, is used to detect operational problems in the cushioning units CUI and CU2. The three accelerometers ACl, AC2, and AC3 together furnish information concerning hunting, wheel quality, track performance, turning, alignment, suspension components, and vibration.

In step IPR of Fig. 6, the microcomputer MCI divides accelerations into different acceleration ranges -5G . . . -G . . . +G . . . +5G, where IG = 32 ft/sec2. An example of such ranges appears in Fig. 7. It keeps a count of significant "shock peaks" or acceleration peaks rom step EXC in each acceleration range. In step IPR of Fig. 6, each time the microcomputer MCI senses a "shock peak" or acceleration peak that fits into one of the ranges, it increments the count in that range by one. The microcomputer MCI keeps running totals of shock peaks that match into each range for a time, such as 30 days, and thus develops a "shock peak distribution" pattern the example of which appears in Fig. 7.

In step SDU, the microcomputer MCI looks for a "shock range distribution unbalance", (SRD unbalance) that is, a distribution unbalance in time accumulations of different ranges of shock maxima. The example of this distribution appears in Fig. 7. In this illustration, the distribution is unbalanced. Here, because the distribution of positive and negative shock peaks among the ranges is expected to be reasonably equal in a cushion, the microcomputer MCI checks for an unbalance in the "shock peak distribution" pattern that exceedε a predetermined permiεεible unbalance. In step SDU, it also quantizes, weights, and records this "shock range distribution unbalance".

In step ERU, the microcomputer MCI looks for an "event range distribution unbalance", (ERD unbalance) that is, a distribution unbalance in time accumulations of different ranges of "events", namely

the aforementioned events f ' ad - άv . This

distribution appears in Fig. 8. Here, the microcomputer MCI divides the events into different event ranges. It keeps a count of "events" in each range. Each time the microcomputer MCI senses an

"event" that fits into one of the ranges, it increments the count in that range by one. The microcomputer MCI keeps running totals of "events" that match into each range for a time, such aε 30 days, and thus develops a "event distribution" pattern an example of which appears in Fig. 8. Because the distribution of positive and negative, i.e. forward and backward, events are expected to be equal in a cushion, the microcomputer MCI in step ERU checks for an unbalance in the "event distribution" pattern that exceeds a predetermined permissible unbalance. It quantizes, weights,

In step RFT, the microcomputer MCI also captures the waveform of each significant impact and examines its rise and fall times. An example of a typical impact acceleration appears in curve A of Fig. 9. The microcomputer MCI measures the rise time and the fall time between percentages such as 10% and 90%.

In step RED, the microcomputer MCI uses the rise times to look for an "rise-time event departure", that is, a significant departure from time accumulations of different rise times that accompany different ranges of "events", namely the aforementioned

events ^t2adt = Δv . This distribution appears in

Fig. 10. Here, the microcomputer MCI measures the acceleration rise time, such as from 0 to 90% of peak, sensed in each significant shock and determines its accompanying event". It divides the events into different event ranges. It memorizes each rise time that accompanies each "event" in each range. The microcomputer MCI keeps running totals of rise times for "events" for a time, such as 30 days, and thus develops a "rise-time event distribution" curve the example of which appearε in Fig. 10. Becauεe the rise times are expected fall within a band around the curve of Fig. 10 to represent a properly operating cushion, the microcomputer MCI, in step RED, checks for a significant "rise-time event departure"

In step RFD, the microcomputer MCI looks for "riεe-time fall-time departures", that is, a significant departure from time accumulations of different rise times that accompany different fall times as shown in Fig. 11. Here, the microcomputer MCI measures the acceleration rise time, such aε from 10% to 90% of peak, sensed in each significant shock and determines itε accompanying fall-time, from 90% of peak to 10%. It memorizes each rise time that accompanies each fall time. The microcomputer MCI keeps running totals of rise times followed by fall times for a period, such as 30 days, and thus develops a "rise-time fall-time" curve an example of which appears in Fig. 11. The rise times and fall times are expected to fall within a band around the curve of Fig. 11 to denote a properly operating cushion. In step RFD, the microcomputer MCI checks for a significant "rise-time fall-time departure" that exceeds a predetermined permissible departure. It quantizes, weights, and records this departure.

In step DBD, the microcomputer MCI estimates whether the impact is a buff or a draft. A departure or off-scale number is evidence of a buff or draft event. Thiε may arise from factε εuch that an oil leak in a cuεhion produceε an air bubble in the cuεhion. A buff impact is generally higher than a draft impact and exhibits a high rise time as shown by curve B of Fig. 9. A draft generally draws out the cushion with lesε of a sudden rise. Typically, a draft impact does not exceed 1.5G. Hence an impact in excess of 1.5G may be estimated aε a buff.

In εtep TQU, the microcomputer MCI then totals the quantized unbalances and departures, each of which represents an individual symptom that may be random or may denote a problem. The evidence of a "shock range distribution unbalance" in one direction or the other counters combined with the evidence of draft or buff in the "rise-time fall-time departure" points toward malfunction in the forward or rear cushion. For example a negative unbalance from the "shock range distribution unbalance" combined with a draft manifestation suggestε a defect in the forward cushion.

As the total quantized unbalances and departures rise, they raise the confidence level that a defect exists in one of the cushions. In step CQT, the microcomputer Mil compares the total of the quantized values with a quantization threshold. In step EQT it asks if the total of the quantized values exceeds the quantization threshold. If the answer is no, in step RMMl the microcomputer MCI returns the process to step PMM. If the answer is yes the microcomputer MCI causes the output device ODl to end a warning signal, together with the location identified by the global position indicator GPl, via the antenna ATl to the central station.

According to another embodiment, the microcomputer MCI, may, in addition to or instead of the antenna signal, set a tag or flag, turn on an LED, or close a switch to notify service personnel that a failure is possible and maintenance is needed. It then returns to step PMM.

The acceleration peaks, the values, the rise times, the fall times, and the acceleration durations each constitutes a measured acceleration characteristic.

For many purposes it is not neccessary to utilize all the data developed by the microcomputer MCI in the flow chart of Fig. 6. According to an embodiment of the invention, one or more of the steps in Fig. 6 is left out. For example, the step DBD may not be needed in an application where the users merely want to detect the exiεtence of an event requiring action from its comparison with the historical data, together with the location of the car RC1, without knowing whether the event is a buff or draft. Accordingly, one or more of the acceleration characteristics need not be^"measured.

According to yet another embodiment of the invention, the microcomputer MCI also determines the performance of the suspensions SU1 and SU2, (or determines only the performance of the suspensions SU1 and SU2) . In such cases it utilizes the output of the accelerometer AC3 in the Z vertical direction orthogonal to the X and Y horizontal directions. For the suspension test, it then operates on the vertically sensed data through the a series of steps corresponding to the steps TST, SIT, PMM, C I, MAA, DDN, DDV, TDA, CTD, EXC, OUT, NOS, of Fig. 3, stepε STO, CMP, SGF, ANL, TRM, (or TMT) of Fig. 5, and more εpecifically εtepε IPR, SDU, ERU, RFT, RED, RFD, TQU, CQT, EQT, SWA, and RMM2 (or RMMl) in Fig. 6. It εpecifically omits εtep DBD which determines buff or draft impacts. The operations are otherwise substantially the same as for the X horizontal data.

During the suspenεion test, in the step DDN of determining the duration in Fig. 3, the microcomputer MCl flags all impacts having a duration less than a predetermined amount, and in step EXC eliminates all such flagged impacts. Flagging durations less than a predetermined value, works as a low pass filter which differentiates performance of the suspensionε SU1 and SU2 from the higher frequency impacts arising from effects such as out of round wheels, truck hunting, and track distortions. Details of step DDN appear in Fig. 6A.

According to yet another embodiment of the invention, the microcomputer MCl also determines the wheel roundnesε (or determineε only the wheel roundness) . for wheel roundnesε it utilizes the output of the accelerometer AC3 in the Z vertical direction orthogonal to the X and Y horizontal directions. For the suspenεion test, it then operateε on the vertically εensed data through the a series of εteps corresponding to the steps TST, SIT, PMM, CWI, MAA, DDN.

During the wheel roundness test, in the step DDN of determining the duration in Fig. 3, the microcomputer MCl divides all impacts less than the one for suspensionε into one of a number of time duration ranges, each of which represents a frequency range. Each range of time durations represents band pass filter. The remaining stepε are each applied εeparately to each of the frequency ranges.

Thus for each of the frequency ranges, the microcomputer MCl proceeds to steps DDV, TDA, CTD, EXC, OUT, NOS, of Fig. 3, stepε STO, CMP, SGF, ANL, TRM, (or TMT) of Fig. 5, and more specifically stepε IPR, SDU, ERU, RFT, RED, RFD, TQU, CQT, EQT, SWA, and RMM2 (or RMMl) in Fig. 6. It specifically omits step DBD which determines buff or draft impacts. The operations are otherwise substantially the same as for the X horizontal data.

According to yet another embodiment of the invention, the microcomputer MCl also determines the the presence or absence of truck hunting (or determines only the performance of the suspensions SU1 and SU2) . In such cases it utilizes the output of the accelerometer AC3 in the lateral, or Y horizontal direction, orthogonal to the X horizontal direction. It then operates on the laterally sensed data through a series of steps corresponding to the steps TST, SIT, PMM, CWI, MAA, DDN in Fig. 3.

During the track hunting operation, in the step DDN of determining the duration in Fig. 3, the microcomputer MCl divides all impacts into one of a number of time durations, each of which representε a frequency range. The remaining εtepε are each applied separately of each of the frequecy ranges.

Thus for each of the frequency ranges, the microcomputer MCl proceeds to steps DDV, TDA, CTD, EXC, OUT, NOS, of Fig. 3, steps STO, CMP, SGF, ANL, TRM, (or TMT) of Fig. 5, and more specifically steps IPR, SDU, ERU, RFT, RED, RFD, TQU, CQT, EQT, SWA, and RMM2 (or RMMl) in Fig. 6. It specifically omits step DBD which determines buff or draft impacts. The operations are otherwise substantially the same as for the X horizontal data. According to another embodiment of the invention, the output device ODl transmits and the antenna ANI propagates the warning signal with the data from the indicator GPl identifying the car RC1 location to a satellite. This appears in Fig. 12 where the satellite is identified as STL. The satellite transmits the signal to a station STA that records this information together with the location of the car RC1 as determined by the global position indicator GPl.

According to the embodiment of the invention in Fig. 1, the antenna ANI and the solar panel SP1 are mounted on the door of the car RC1. According to the embodiment shown in Fig. 13, the antenna ANI and the solar panel are mounted on the top of the car RC1. According to the embodiment of Fig. 14, the unit MUl is mounted on the rear of the car RC1 and the antenna ANI and the solar panel SP1 are mounted on the unit MUl.

According to another embodiment of the invention, the outputs of others of the accelerometers ACl, AC2, and AC3 are used separately or combined as components in the microcomputer MCl. To determine wheel quality, the microcomputer MCl utilizes the outputs of accelerometer AC3 along the Z axis, i.e. the vertical axis. That is impacts in the vertical direction indicate out of round conditions. The process in the flow chart of fig. 6 is used for this and other purposes. For "truck hunting" i.e. the sway of the pivoted undercarriages that each carry four wheels of the car RC1, the y axiε and the rotational effects are calculated by the microcomputer MCl.

As stated, in step PPM, the microcomputer MCl also turns on the global position indicator GPl for a predetermined interval sufficient to capture position data, and then sets the indicator GPl to turn on at periodic intervals, such as 30 seconds. Fig. 3A illustrates details of a εtep Sll after the step PMM. The periodic capture is important because it can take up to 15 minuteε for a GPS or Loran unit to get an accurate position report. The embodiment recognizes that the time delay is primarily a function of how long it has been since the last report was made from that unit, how far the unit has moved since its last report, and its specific location.

However it is often important to know the precise physical location of the event. If there is, for example, a two minute delay, this can translate to over 2 miles on a train or truck travelling at 60 miles per hour.

This embodiment is set to turn on the indicator GPl at small intervals such as 30 seconds, and preferable even smaller, to get an accurate position update. (However, longer periodε up to two minutes are possible where acceptable.) Then when an event occurs, the lag between the event and the accurate position report is minimized. The embodiment allows very precise location reporting without the power consumption required if the indicator GPl were left on continuously.

According to another embodiment of the invention, the system uses a GPS station at a precisely known location to serve as a reference. It is read periodically as a central base station. According to another embodiment of the invention, the solar panel SPl and the antenna ANI are integrated into one unit.

Yet another embodiment of the invention serves also for monitoring other operational aspectε within a vehicle or closed container and relaying that information, with geographic position data, back to a central reporting station. Typical examples include temperature and impacts in a vehicle or cargo container.

An embodiment involveε determining or estimating if variouε conεtituentε of a rail- road cargo-carrying car are functioning properly.

According to an embodiment the operation of Figs. 5 and 6 stores only the data that exceeds the second threshold.

The invention obtains desired information quickly, and with precise vehicle or container location information. In thiε way, inεpectors may be immediately dispatched to a vehicle recently receiving a suspected impact, or incorrect temperature report to determine the cause for the anomalous report.

While embodiments of the invention have been described in detail, it will be evident to thoεe skilled in the art that the invention may be embodied otherwise without departing from its spirit and scope.

Claims

What is claimed is:

1. A method of monitoring events in a vehicle, comprising:

the step of producing electrical outputs repreεentative of events in the vehicle; and

the step of comparing characteristicε of one event with characteriεtics of other events accumulated over a given period of time and determining departures of a given extent from the other characteristicε as an indication of a significant event.

the step of sending a warning in response to the indication with periodically captured position data.

2. A method as in claim 1, further compriεing the εtep of capturing poεition data non- continuouεly and periodically in periods less than 2 minutes;

3. A method as in claim 1, wherein the step of capturing includes producing a position signal showing the position of the vehicle with a global position indicator.

4. A monitoring method as in claim 1, wherein the events are impacts on the vehicle;

the step of producing electrical outputs includes producing outputs representative of acceleration characteristics during an impact with an accelerometer system mounted on said vehicle; and

the step of comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts and determining departures of a given extent from other characteristics as a failure indication.

5. A monitoring method as in claim 3, wherein the events are impacts on the vehicle;

the step of comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impactε and determining departures of a given extent from other characteristics as a failure indication.

6. A monitoring method as in claim 2, wherein said comparing step includes counting the number of characteristics that fall into given ranges of characteristicε in a given time and determining departureε from given distribution of counts among the ranges.

7. A monitoring method as in claim 1, wherein one of said characteristics is the acceleration peak of the impacts.

8. A monitoring method as in claim 1, wherein one of said characteristicε is

9. A monitoring method as in claim 3, wherein one of εaid characteristics is

10. A monitoring method as in claim 2, wherein one of said characteristics is the rise time of the impact.

11. A monitoring method as in claim 10, wherein said comparing step compares rise times with

12. A monitoring method as in claim 10, wherein said comparing step includes comparing rise times with fall times.

13. A monitoring method as in claim 10, wherein said comparing step includeε combining a plurality of departureε from a plurality characteristics to obtain an indication of failure.

14. A monitoring method as in claim 4, wherein said comparing step includes combining the relationship of acceleration in the vertical direction as an indication of suεpenεion condition.

15. A monitoring method as in claim 4, wherein

the εtep of producing electrical outputs includes producing outputs representative of acceleration characteristics during an impact in a vertical direction of the vehicle, and

the step of comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts in the vetical direction.

16. A monitoring method as in claim 4, wherein

the step of producing electrical outputs includes producing outputs representative of acceleration characteristics during an impact in a plurality of orthogonal directions of the vehicle, and

the step of comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts in the plurality of orthogonal directions.

17. A system for monitoring events in a vehicle, comprising:

an electrical output generator for producing outputs representative of events in the vehicle;

meanε for capturing the position of the vehicle periodically and non-continuously at periods less than 2 minutes; and

a warning signaller responsive to the indication and to said capturing means for producing singnals with both the warning and the position.

18. A system as in claim 17, further comprising means for comparing characteristics of one event with characteristicε of other eventε accumulated over a given period of time and determining departures of a given extent from the other characteriεtics as an indication of a εignificant event; and

19. A system as in claim 17, wherein said warning signaller includes a global position indicator.

20. A monitoring system as in claim 18, wherein the events are impacts on the vehicle;

the electrical output generator includes an accelerometer arrangement mounted on the vehicle and the outputs are representative of acceleration characteristics during an impact; and

the means for comparing and determining includes means for comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts and determining departures of a given extent from other characteristics as a failure indication.

21. A monitoring system as in claim 19, wherein the events are impacts on the vehicle;

the step of producing electrical outputs includes producing outputs representative of acceleration characteristics during an impact with an accelerometer εyεtem mounted on εaid vehicle; and

the εtep of comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts and determining departures of a given extent from other characteriεticε as a failure indication.

22. A monitoring syεte as in claim 18, wherein said comparing means includes counting the number of characteristics that fall into given ranges of characteristics in a given time and determining departures from given distribution of counts among the ranges.

23. A monitoring system as in claim 18, wherein one of said characteristics is the acceleration peak of the impacts.

24. A monitoring system as in claim 17, wherein one of said characteristics is

25. A monitoring system as in claim 19, wherein one of εaid characteristics is

26. A monitoring system as in claim 17, wherein one of said characteristicε is the riεe time of the impact.

27. A monitoring εystem as in claim 18, wherein said comparing means compares rise times with

28. A monitoring syεtem aε in claim 18, wherein εaid comparing means includeε comparing rise times with fall times.

29. A monitoring system as in claim 18, wherein said comparing means includes combining a plurality of departures from a plurality characteristics to obtain an indication of failure.

30. A monitoring method as in claim 20, wherein said comparing means includes combining the relationship of vertical acceleration as indication of suspension condition.

31. A monitoring system as in claim 20, wherein

the electrical output generator includes an accelerometer arrangement and the outputs are representative of acceleration characteristics during an impact in a lateral direction of the vehicle, and

the means for comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteriεtics of other impacts in the lateral direction.

32. A monitoring εystem as in claiπ 20, wherein

the electrical output generator includes an accelerometer arrangement and the outputs are representative of acceleration characteristics during an impact in a plurality of orthogonal directions of the vehicle, and

the means for comparing and determining includes comparing acceleration characteristics of one impact with accumulated acceleration characteristics of other impacts in the plurality of orthogonal directions.