US3258744A - Vehicle traffic control system - Google Patents

Description

June 28, 1966 J. H. AUER, JR

VEHICLE TRAFFIC CONTROL SYSTEM Filed Aug. 24, 1961 Z Sheets-Sheet 1 FIGI. I} -0ET.

1 I i l l l l TRAFFlC I SIGNAL a W I l I 1 i E i F 6 I 1 0w 3 7 DE DET. COMPUTER ----JDET./

| L SIGNAL 0s CONTROLLER 1- l FAILURE L J TIMER INVENTOR.

AUER JR.

HIS ATTORNEY June 28, 1966 J. H. AUER, JR I 3,258,744

VEHICLE TRAFFIC CONTROL SYSTEM Filed Aug. 24, 1961 I L- Sheets-5heet VEHICLE PRESENCE DETECTOR FIG. 3. I4

I2 /H IO SQUND FREE v HICLE PAVEEMENT E LE"TION REFL CTION djAMPLIF|ER-- PULsE RUNNING F GATE GENERATOR MULTIVIBRATOR GENERATOR GENERATOR vEI-IICLE REFLECTION AM5EI I5R 5 6\ ,19 /2O RECTIFIER- CONDITION AMPLIFIER- 4 FILTER l8 FLIP-FLOP DETECTION PAVEMENT L REFLECTION RELAY GATED CONTROL AMPLIFIER CIRCUIT I RN FIG. 5. Efi

WAVEFORM DIAGRAMS A TUBE v2 CATHOOE LE I'QE VOLTAGE v I I l I l B. TUBE v2 PI ATE r I CURREN I ;I v TUBE v3 PLATE \I CURRENT I I H D. RELAY TI l I I l E RELAY T2 m I 1//LCUTOFF A I I I 1 F. TUBE v4 GRID 'I/ I i VOLTAGE I G. TUBE v4 PLATE F1 CURRENT H. RELAY A2 L II. RELA PCI A L K RELAY AI /1 IVVEN'TUR. JHAUER JR.

HIS ATTORNEY June 28, 1966 J. H. AUER, JR 3,258,744

VEHICLE TRAFFIC CONTROL SYSTEM Filed Aug. 24, 1961 3 Sheets-Sheet 3 LLI 5 2 8 o E s P 5 O o O O O O O 8 9 8 2 8 9 g INVENTOR. JHAUER JR. 1- BY HIS ATTORNEY 3,258,744 VEHICLE TRAFFIC CONTROL SYSTEM John H. Auer, Jr., Rochester, N.Y., assignor to General Signal Corporation, a corporation of New York Filed Aug. 24, 1961, Ser. No. 133,616 14 Claims. (Cl. 340-37) This invention relates to a system for the control of trafiic signals, and more particularly pertains to a system wherein the signals are controlled in accordance with traffic congestion.

This application is a continuation-in-part of my copending application Serial No. 78,410, filed December 27, 1960, now US. Patent No. 3,233,084.

Fixed time signals for the control of vehicular trafiic are in wide use and in many instances they provide adequate control of traific. It has been found, however, that the traffic-handling capacity of highways may often be increased appreciably by providing vehicle-actuated control over the signals. In general, this is accomplished by providing a vehicle detector along one or more approaches to an intersection and then controlling the length of the operating cycle, and/or also the splitting of the cycle between the different directions of travel in accordance with the output of the various vehicle detectors.

In some of the presently used, vehicle-actuated traffic control systems, preference is usually given to traffic moving along a main artery by providing that the signal thereon will provide a continuous green aspect until the presence of traific on an intersecting side street has been detected, at which time the trafiic on the side street is given a green signal for at least some minimum interval of time. In some of these systems, the detection of only a single vehicle approaching the intersection from a side street will control the signal system to give a green aspect for that side street, provided only that there has until then been a green signal displayed for the main artery for at least some minimum interval of time. In still other systems, the presence of only a single vehicle on the side street may not immediately cause a green signal to be dis played for traific on the side street, but instead the amount of time that elapses before a green signal is displayed for such side street is dependent upon the number of vehicles detected as approaching the intersection along such side street, with the waiting interval being, of course, reduced when the traffic is at a higher level. Even in such a system, however, even the single vehicle on the side street will eventually be effective to cause a green signal to be displayed for it, but where the traffic on the side street is light, then frequently a longer time interval must elapse before it receives a green signal.

Generally, in these systems of the prior art, it is provided that the length of time throughout which the side street will continue to receive a green signal will be dependent upon the number of vehicles which have passed the detector on that side street. Thus as traffic becomes heavier, more vehicle counts will be produced from the vehicle detectors and, in response thereto, successive increments will be added to the green time for the side street up to some predetermined maximum value, at which time the signal will go to red for the side street and traffic will again be permitted to move along the main artery. In some of the more elaborate systems, vehicle detectors are placed as well along one or more lanes of the main artery so that the amount of trafiic on the main artery can as well be gauged and thereby make it possible to control the amount of green time which is displayed by the signals for traffic moving along the main artery.

Although these prior art systems just described are frequently more effective than fixed time control of the tratlic signals, their effectiveness often decreases at times ice when their proper functioning becomes more critical. Thus, it is a characteristic of these systems that the vehicle detectors employed are not actually vehicle presence detectors but merely sense the passage of vehicles and thus provide the same kind of output for each passing vehicle, regardless of its length or of its speed. The detector means frequently used is of the type in which a treadle is mounted fiush with the roadway surface and is actuated by the wheels of the vehicles as they roll over the treadle. Such a detector thus generally provides two output pulses for each passing auto, but will produce four or more successive actuations for a single passing truck so that the number of output pulses produced by such a detector is not a clear indication of the number of vehicles passing the detector location. Moreover, such a detetcor cannot provide any information as to the speed of moving vehicles. It is recognized that the successive pulses produced by the front and rear wheels of a car are more closely spaced when that car is moving quickly than when it is moving slowly. However, the difiicultics that exist in determining whether two successive pulses have been produced, on the one hand, by the successive wheels of the same vehicle, or, on the other hand, by the rear wheel of one vehicle and the front wheel of another, make it unfeasible to produce output signals from such a detector which will represent more than merely the number of vehicle actuations occurring in a given time.

Because of these shortcomings, which are by no means limited to the treadle type detectors but are indeed inherent in all detectors not of the presence detecting type, their effectiveness in a system intended to evaluate actual traiiic congestion is apparent only under conditions of relatively light traffic congestion. At such times, trafiic can flow quite freely, and the successive veh cles in a line of trafiic at an approach to an intersection will continue to pass the detector so that its rate of output counts increases as the amount of traffic increases. An increment of time may be added to the green phase of the traflic signal upon detection of an increase in traffic with the result that the increased trafiic can quite readily be accommodated by being permitted to cross the intersection. Placement of the vehicle detectors at a substantial distance in approach of the intersection ensures that a line-up of a number of cars waiting for a green signal will still not result in a blockage all the way back to the location of the detector, and thus it will be possible for the newly arriving cars to be detected by the vehicle detector and have their approach to the intersection also recognized so that they too will be influential in determining the amount of green time that will be displayed.

Under extremely heavy traffic conditions, however, traffic necessarily moves more slowly and, additionally, the line of vehicles waiting at an intersection for a green signal will become so long that it will soon extend back to the detector location. Under these circumstances, there are vehicles adjacent to or within the detection zone of the detector which are not moving at all for subtantial periods of time or which move very slowly even when in motion. This means that a detector which is responsive only to the passage of vehicles and ignores their length and velocity will produce only very few output counts in any given measuring interval and will thus indicate a very low volume of traflic at times when trafiic volume is actually at a peak. In other words, under these most critical conditions, when it should he recognized by the system that traflic congction is at its worst, the signal sysem instead is detecting only a very low volume of tralhc. and the signals are then being so controlled that trailic congestion is made more acute rather than being aided.

Another related situation which frequently results is that movement of lltllllC under high truiiic load conditions may be substantially out of phase with the operation of the signal system. Most drivers have experienced the situation where their forward progress actually occurs principally at times when the signal ahead is displaying a stop indication but where their progress is practically nil when the signal ahead is giving a proceed indication. This situation is experienced mostly by vehicles which are at a substantial distance from the signal location and which may thus be in the vicinity of the vehicle detector. This situation further aggravates the problem, particularly when the system is of the kind in which, as previously mentioned the amount of green time displayed for a particular approach is a function of the amount of trafiic detected on the approach during the green phase. Thus, during such green phase, trafiic in the vicinity of the detector may be virtually at a standstill and this will result in a relatively short green time for the signal since the system has in fact detected only a small number of vehicles as having passed the detector location.

With these consideraitons in mind, it is apparent that a vehicle-actuated traffic control system cannot properly be operated by vehicle detectors which are responsive merely to the passage of vehicles since traffic volume (the number of vehicles counted within a unit of time) tends to increase with density (number of vehicles per mile) and with occupancy (percentage of road occupied) only for values of density or congestion which are relatively low. In other words, for conditions of relatively light traffic, trafiic volume does tend to vary at about the same rate as vehicle density and vehicle occupancy; however, volume reaches a peak long before density and occupancy reach their maximum values as is evidenced by the fact that volume will go to substantially zero as density and occupancy near one hundred percent. Necessarily, therefore, traflic volume is not a reliable guide to the amount of traffic which must be handled.

It is contemplated by the present invention, therefore, to provide a traffic signal control system in which the vehicle detector used on each approach to a traffic signal is of the presence-detecting type, i.e. it is distinctively operated when a vehicle enters a detection zone defined by the detector and remains in such operated condition for so long as is required for the vehicle to pass through the detection zone. Preferably, the detector employed is of the type shown in the patent application of Kendall et al., Scr. No. 808,736, filed April 24, 1959, now US. Pat. No. 3,042,303 issued July 3, 1962, since such a detector embodies a unique interlocking which prevents any momentary loss of vehicle detection even under the most unfavorable conditions. Since a detector is deemed to be much preferred over other types of so-called presence detectors, especially those which merely only rely upon the reflection or radiation of energy from a passing vehicle since it has been discovered that 'the energy thus obtained may momentarily be lost, with the result that the signals produced are not at all times truly representative of vehicle presence. It is further contemplated that the present invention will utilize the presence signal output of the vehicle detector, in providing another different signal which will represent occupancy and that the different occupancy signals so produced for the different approaches will control both the overall cycle time and also the splitting of the cycle turn into respective phases whose relative durations will accord with the respectively different occupancy signals. It is also proposed that the control system be so organized that it can readily be adapted for use with a fixed time controller and so adapted, moreover, that the fixed time controller Will automatically come into use upon the occurrence of a failure in the apparatus of this invention.

It is, therefore, an object of this invention to provide a vehicle actuated traffic control system in which the timing of signals for the difIcrent approaches to an intersection is governed by the congestion on the approach to the intersection as measured by vehicle occupancy rather than by a measurement of vehicle volume.

It is another object of this invention to provide a system for the control of traffic signals at an intersection in which the relative green times of the several approaches are timed to be a function of the traffic occupancy on the respective approaches.

It is an object of this invention to provide a system for traffic signals in which the timing of the signals is controlled in accordance with the amount of traffic to be handled over the respective approaches to an intersection, with the measurement of trafiic conditions being accurately determined over its entire range of variation.

It is another object of this invention to provide a control system for a traffic light in which the operating times of the respective phases may each be varied independently in accordance with the amount of traffic approaching an intersection from the respectively different directions.

It is another object of this invention to provide a control system for traffic in which the signals shall be controlled according to traffic congestion but shall also, in the event of any failure of the apparatus which measures the traffic congestion, automatically provide fixed time control for the signals so that the signals will continue to operate in the event of such failure.

It is still another object of this invention to provide a control system for traffic signals in which the system is at all times self-cycling, but wherein the relative times of the proceed indications for the different approaches are proportional to traffic conditions on such respective approaches.

An additional object of this invention is to provide a control system for traffic signals in which the occupancy of the highway in approach of the signal is measured and where such measurement is made well in advance of the intersection itself so that traffic does not ordinarily back up to the point of measurement and thereby give erroneous readings of road occupancy.

Further objects, purposes, and characteristic features of this invention will in part be clear from the accompanying drawings and will in part become clear as the description of the invention progresses.

In describing this invention in detail reference Will be made to the accompanying drawings in which like reference characters designate corresponding parts throughout the several views and in which:

FIG. 1 illustrates diagrammatically a typical intersection controlled by one or more traffic signals and illustrates the use of vehicle detectors on the different approaches to provide a measurement of traffic congestion for control of the signals;

FIGS. 2A and 2B illustrate a typical mounting of the transducers of a vehicle detector of the type preferred for use in the present invention and further illustrate the way in which the propagation path of the transmitted energy varies in accordance with whether or not a V6- hicle is within the detection zone;

FIG. 3 is a block diagram of a preferred vehicle detector of the presence indicating type;

FIG. 4 is a circuit diagram of a typical embodiment of the present invention;

FIG. 4A illustrates a modified form of the circuit of FIG. 4; and

FIG. 5 is a wave-form diagram illustrating typical voltage and current wave forms appearing at various places in the circuit organization of FIG. 4.

For the purpose of simplifying the illustrations and facilitating the explanation of this invention, the various parts and circuits have been shown diagrammatically and certain conventional elements have been shown in block form since the drawings have been made more with the purpose of making it easy to understand the invention rather than to illustrate the specific construction and arrangement of parts that might be used in practice. The symbol (Ii-l) and the symbol for ground connection indicate connections to the opposite terminals of source of electrical energy, At various places in the circuit diagram of FIG. 4, specific voltage levels have been illustrated, but these have been shown merely for the purpose of facilitating the explanation of the invention and it is understood that the invention is in no way limited to the use of voltages of the magnitude shown.

Described briefly, it is contemplated by the present invention to provide a vehicle detector which is capable of registering the presence of each vehicle approaching the controlled intersection. In other words, with such detector an output signal is provided for each vehicle whose duration corresponds to the length of time required for the vehicle to pass the detection location. Such detector is considered to be preferable over the type of detector which merely counts axles of passing vehicles since the latter as already mentioned does not ordinarily reflect truly the state of traffic congestion of the roadway. Each detector is mounted considerably in advance of the intersection so that any vehicles which are waiting for the traffic signal will ordinarily not be stopped while within the detection zone of the vehicle detector since, whenever this happens, the vehicle detector will necessarily indicate maximum highway congestion. With the detector located only three or so car lengths in advance of the intersection, maximum congestion would frequently be recorded at times when there were only a few cars present and erroneous data would then be provided. However, if the vehicle detector is a substantial distance in advance of the intersection, then any queue which forms and results in a vehicle being stopped within the detection zone of the vehicle detector will properly indicate a high degree of congestion. The output signal obtained from the vehicle detector does not itself provide a direct measurement of occupancy, but occupancy may nevertheless be conveniently ascertained therefrom as will be shown. It is further proposed that the duration of the green phase for each direction of travel be a function of the occupancy measured for that direction of traffic, with the sig nal being so controlled that the green phase will vary from a preselected minimum duration for minimum occupancy to some predetermined maximum duration when occupancy is measured as being at a high value.

Referring to FIG. I, a typical intersection of two streets is shown. It is assumed that there is two-way traffic on each street and that trafiic in the respective directions is controlled by trafiic signal 6 which is shown positioned in the center of the intersection, although it will readily be understood that multiple signals may be placed upon the various corners of the intersection as well. For each of the approach lanes governed by the signal 6 there is a detector such as the detector DE associated with the East approach lane, and this detector defines a detection zone over the lane of the East approach which will be traversed by each vehicle as it appreaches the intersection.

The vehicle presence detector used may be of the kind disclosed and claimed in the prior copending applications of H. C. Kendall et al., Ser. No. 808,736 filed April 24, 1959, now US. Patent 3,042,303 is ued July 3, 1962, or J. H. Auer, Jr., Ser. No. 820,325, filed June 15, 1959, now US. Patent 3,045,909 issued July 24, 1962. In both such prior applications, the vehicle detector is disclosed as one wherein a beam of energy is transmitted across the path of each vehicle so that it impinges upon each vehicle and is reflected back toward a receiving transducer for as long as it takes the vehicle to pass through the beam. Preferably, in practice, the beam of energy is directed downwardly so that it normally impinges upon the pavement as shown in FIG. 2A but impinges instead upon a vehicle as long as it is within the detection zone as shown in FIG. 23. It is assumed for the purposes of illustration in FIG. I, that the detector is of this latter type so that a generally circular detection zone is defined as diagrammatically shown in FIG. 1. It will also be understood that if there are two or more lanes which are to be monitored, the detection zone can be mad: broader so that a vehicle approaching the intersection by travelling in either of the adjacent lanes will have its presence detected. It is generally preferred, however, that in such an instance separate detectors be employed, one for each of the several lanes so that vehicles which are generally abreast of each other and travelling toward the intersection will be separately registered. Quite clearly, the latter arrangement will provide a more accurate determination of highway occupancy. Still another alternative is to detect vehicle presence in only one of the several lanes, and then assume that the occupancy determined therefrom is representative, as well of occupancy in the adjacent lanes.

FIG. 3 illustrates in block diagram form a vehicle presence detector of the kind disclosed in the previously mentioned application Ser. No. 820,325, now US. Patent 3,045,909 issued July 24, 1962. This detector, in its preferred form, employs sound energy which is transmitted in the form of discrete pulses. In FIG. 3 the apparatus provided for generating the repetitive sound pulses comprises the free-running multivibrator 10 whose frequency of operation establishes the pulse repetition rate. For each cycle of its operation, the multivibrator 10 applies a triggering pulse to the sound pulse generator 11. The generator 11 then produces a brief pulse of ultra sonic frequency energy which is amplified by amplifier 12 and applied to the transmitting transducer T.

The round-trip progagatfon time of a reflection pulse is affected considerably by whether or not there is a vehicle present within the sound beam. When no vehicle is present, each sound pulse has the maximum propagation time, but when a vehicle is present the propagation time is considerably reduced. Electronic gating circuits are employed and each demarcates a succzssive time interval. The first of these gating circuits, the vehicle reflection gate generator 13, demarcates a time interval which encompasses that period of time after the transmission of each sound pulse when a vehicle reflection can be expected to be received. Another subsequent time interval is demarcated by the pavement reflection gate generator 14 and encompasses the interval of time during which a pavement reflection can be expected to be received.

The reflection pulses are all amplified by amplifier l5, and after being rectified and filtered by the rectifierfilter 16, are applied to both the vehicle reflection gated amplifier 17 and a pavement reflection gated amplifier 18. These two amplifiers are respectively gated by the voltages derived from the gate generators l3 and 14 previously referred to. As a result, when no vehicle is present, each transmitted sound pulse results in a corresponding output from the pavement reflection gating amplifier 18, and this output is then applied to one input of flip-flop 19. Similarly, when a vehicle is within the sound beam, each transmitted sound pulse produces an output from the gating amplifier 17, and this is then applied to the other input of flip-flop l9.

Flip-flop 19 is, therefore. in one of its two stable states whenever no vehicle is present so that it receives successive input pulses from amplifier l8. Flip-flop 19 is operated to its opposite state, however, when a vehicle is within the sound beam so that it receives successive input pulses from amplifier 17. Condition detector 20 is connected to flip-flop l9 and senses which of its two states tlip flop 19 is in at any time. Whenever condition detector 20 senses that flip-flop 19 is in that condition which it assumes whenever it is constantly receiving output pulses from amplifier 18, it then acts upon relay control circuit Zll to cause relay RN to be dropped away so that front contact 22 of this relay is open. On the other hand. when condition detector 20 is in the opposite of its two states by reason of having sensed that ili flop 19 is in the condition it necessarily assumes whenever it receives successive pulses from amplifier 17, then relay control circuIt 21 is operated to the condition where relay RN is picked up. At such time, from contact 22 of relay RN is closed.

From this description, it can be seen that the vehicle detector shown in FlG. 3 is of the type which is a P euce" detector in that relay RN is picked up throttghout the time that a vehicle is detected as being within the sound beam. The various component values and time constants associated with this detector are so chosen that the response and release time for relay RN will be substantially identical. This is done so that the picked up time of relay RN will tend to be closely proportional to the length of time that the vehicle is within the sound beam.

Another distinctive characteristic of the vehicle detector of FIG. 3 is that it provides a high degree of dis crimination against spurious objects. More specifically, in order for a vehicle to be detected, it is first necessary that the normally reoeived pavement reflections be no longer received and that concurrently therewith vehicle reflection pulses be received. Before the apparatus can be restored to its normal condition so that it can thereafter again detect a subsequent vehicle, it is necessary that the vehicle reflections cease and that the pavement reflections again be restored. These multiple requirements ensure that only a vehicle will ordinarily provide operation of the detector relay RN. Moreover, a convertible automobile having sound reflecting surfaces only at its front hood and rear deck portions cannot possibly be counted as two separate vehicles, since the absence of pavement reflection pulses when the cloth top is within the sound beam prevents the detector relay RN from dropping away. These characteristics of the vehicle detector of FIG. 3 are fully set forth in the prior application Ser. No. 808,736, now Patent No. 3,042,303, referred to previously. In addition, the prior application of J. H. Auer, Jr., Ser. No. 820,325, filed June 15, 1959, now Patent No. 3,045,909 may be referred to for a detailed description of the mode of operation of the flipfiop 19, condition detector 20, and relay control circuit 21.

Although the vehicle detector of FIG. 3 is disclosed as being of the pulsed ultrasonic type, it is to be emphasized that it is by no means essential that a vehicle detector of this particular type be used. It is only necessary that the detector generally be of the type which will provide an output for each vehicle whose duration is proportional to the length of time that the vehicle requires to pass a given point, i.e. in this case, to pass through the detection zone defined by the beam of repetitive sound pulses.

The output of each of the several detectors monitoring traffic flow toward the intersection over the various highway segments is applied to computer apparatus, shown in detail in FIG. 4 and in block diagram form in FIGURE 1. The function of this computer apparatus whose operation will be described in detail is to determine cycle duration and cycle split between the respective phases for signal 6. More specifically, the computing apparatus includes an East-West lane occupancy computer 7A, a North-South lane occupancy computer 7B, and selector relays 7C which are, themselves, selectively actuated by signal controller 8. Each of the lane occupancy computers 7A and 7B is responsive to one or more vehicle presence detectors and each, as will subsequently be described in detail, provides an output signal control manifestation which is a function of the lane occupancy measured in either the North-South or East-West approaches to the intersection. The selector relays 7C determine which of the lane occupancy computers 7A or 7B shall be electrically coupled to the controller 8 at any time. In the illustrated embodiment of the invention, the relay selection is such that the particular lane occupancy computer which is connected to the controller 8 at any time corresponds to the particular direction of Il'tlillC then receiving a procccd indication. in other words, throughout the time that controller 8 is controlling the trallic signals to display :1 proceed indication for the East-West trallic directions, the controller is responsive to the output of East-West lane occupancy measuring apparatus 7A, thereby permitting the duration of this proceed indication to be dependent upon the lane occupancy then being measured on the East-West approaches to the intersection. Associated with the controller 8 is a failure timer FT. As will subsequently be described, if there should be any failure in the computing apparatus or in any associated vehicle detector so that controller 8 is not operated to advance the signal system for a length of time greater than some predetermined maximum time, the failure timer FT will recognize this situation and will, in effect, disconnect controller 8 from the computing apparatus and cause the controller 8 to become self-operating, dependent entirely then upon its own internal timing mechanism. When this happens, control of the signal is then no longer according to traflic conditions; nevertheless, the signal will periodical- 1y change its indications and thus prevent a complete stoppage of traffic in one direction as would ordinarily occur.

Referring to the detailed circuit of FIG. 4 relay RN of FIG. 3 is here shown as being the relay associated with the vehicle detector DN which, as shown in FIG. 1, is the detector monitoring the approach of vehicles towards the intersection over the North approach. In a similar manner, the detector DS for the South approach controls the actuation of relay RS, and detectors DE and DW on the East and West approaches are associated with relays RE and RW, respectively. Normally, each vehicle detector maintains the associated relay in a dropped-away condition but controls the relay to pick up for as long as a vehicle occupies the associated detection zone.

Each detector and corresponding relay controls the input signal to a respective cathode follower tube such as tube VN which is associated with detector DN and relay RN. When relay RN is dropped away, the grid of the tube VN is connected through resistor 24 and through back contact 25 of relay RN to an adjustable tap on potentiometer 26. When relay RN is picked up, the grid of tube VN is then connected instead through resistor 24 and through front contact 25 of relay RN to the movable tap of potentiometer 27. If relay RN were to stay in its dropped-away condition for a relatively long time, capacitor 28 connected from the grid of tube VN to ground would become charged to the level of voltage which appears at the tap of potentiometer 26. Similarly, if relay RN were to be picked up for a long interval, the charge on capacitor 28 would eventually equal the voltage appearing at the tap of potentiometer 27. In practice, these potentiometers are so adjusted that the voltage at the tap of potentiometer 26 is appreciably less than at the tap of potentiometer 27. For this reason, the amplitude of the voltage that does actually appear across the terminals of capacitor 28 is a function of the proportionate amount of time that relay RN is picked up. As is described in considerable detail in the prior application of Auer, Ser. No. 78,410 filed December 27, 1960, with respect to which the present application is a continuation-in-part the circuit organization just described produces a voltage across capacitor 28 that varies with traflic conditions in such a manner that it quite closely approximates road occupancy. i.e. the percentage of pavement occupied. It has been shown in practice that this measure of occupancy is closely correlated to traftic congestion.

The voltage across capacitor 28 determines the platecathode current of tube VN and thus determines also the amplitude of the output voltage appearing across the cathode resistor 29 of this tube. Because of the cathode follower action, this output voltage is quite close in amplitude to that appearing across capacitor 28. A similar circuit organization is provided for each of the other vehicle detectors and detector relay s.

It is assumed that the system shown in H0. 1 contemplates the use of a two-phasc signal cycle. More summing points S1 rather than their average value.

specifically, trafiic on both the North and South approaches simultaneously receives a green signal and thereafter traffic on the East and West approaches similarly is given green signals simultaneously. It is further assumed that each signal advances from green to amber, to red, and back to green again. Under such conditions, it is desirable to combine occupancy measurements for respectively opposite directions, i.e. traflic occupancy as measured on both the North and South approaches may be used to control the duration of the green phase displayed simultaneously to trafiic on the North and South approaches. To accomplish this, it has been found in some cases desirable to obtain an average of the occupancies measured for the two opposite approaches and use this average to determine the duration of the green phase. Thus, both the cathode voltage of tube VN (representing occupancy on the North approach) and the cathode voltage of cathode follower tube VS (repressenting occupancy on the South approach) are applied through respective summing resistors to a common summing point 51 in FIG. 4. The voltage at this summing point St is therefore an average of that supplied by the respective North and South approaches.

In other cases, it may be desirable to replace the summing resistors with diodes as shown in FIG. 4A so as to apply the higher of the two voltages to the common In this way, the North or South approach having the greatest occupancy provides the factor of control. Similar diodes may be provided for the output circuits of cathode follower tubes VB and VW in place of the resistors shown in FIG. 4A. In certain other instances, it may be desirable to use diodes for the two cathode followers associated with one direction of traffic and summing resistors for the cathode follower tubes associated with the other direction of tratfic. It will be apparent that the choice is entirely in accordance with the particular results desired to be achieved in any instance.

In a manner similar to that just described, traffic on the East and West approaches which is monitored respectively by the detectors DE and DW, results in the appearance of an average voltage at the common summing point 82 which then represents the average occupancy conditions for the East-West approaches.

Before considering how these voltages representing occupancy for the respective North-South, East-West directions of traffic will be utilized in the circuit of FIG. 4, it is believed desirable to describe in a general manner what is sought to be accomplished by control of the signal system in accordance with occupancy. Thus, it is proposed that each green phase vary in duration over selected limits in accordance with traffic occupancy on the respective approaches receiving the green signal, with the duration being some predetermined minimum for low values of occupancy and increasing to some predetermined higher value as occupancy increases to a maximum. It is proposed that each of the green phases will be independently controllable in this manner. Mathematically, therefore, the duration of a green phase may be represented by the, formula t=A +BX in which A and B are constants which may be adjusted to predetermined values and in which X is proportional to the occupancy of traffic on the respective approach governed by that phase and may be expressed as a percentage of total, or 100%, occupancy. The value of A fixes the minimum duration of the green phase which occurs under zero occupancy conditions. The value of B fixes the rate at which the duration of the green phase increases with increascs in occupancy and also establishes the maximum duration of the green phase occurring when occupancy approaches 100%. A linear relation between measured occupancy and time t as suggested by the above equation is believed to be suitable in most instances, but it will be appreciated that the invention embraces as well the concept of a nonlinear relationship. Assuming typical values, A may be 15 seconds and B may be 30 seconds. In that event, when there is no East-West traflic, i.e. zero occupancy, the length of the green phase for that direction of trafiic will be 15 seconds but will increase as occupancy increases so that for an occupancy of the time of the green phase will become 45 seconds.

Since the respective green phases are independently controllable, it follows that although the green phase for the East-West direction may go as high as 45 seconds under high occupancy conditions, the time of the green phase for the North-South direction can still remain at a minimum as long as occupancy for the North-South approaches remains low. If the constants A and B for this North-South phase are the same as those just suggested in the East-West phases, then the length of the green phase for North-South may remain at 15 seconds even though the length of the green phase for the East- West trafiic may vary upwardly toward a 45 second interval.

One of the advantages of having the different green phases independently variable in the manner just described is that under light traffic conditions the signal cycle is short so that no operator of a vehicle experiences a long delay. Under the assumed conditions just described, and also assumingthat occupancy conditions are very light so that there is only an occasional vehicle in either direction, no operator of a vehicle will experience a delay of more than approximately 15 seconds. Of course, as traffic builds up on one street while that on the other street remains light, the split in the signal cycle automatically becomes unbalanced so as to favor the street with the heavier tratiic. Moreover, if trafiic on both streets becomes heavy, the signal cycle split may remain balanced but the cycle length then is increased for both the North-South and East-West directions. One result of this is that during any given measuring interval there is less total amber time during which traffic flow is restricted in both directions and also traffic flows more smoothly and with higher average speed since there are less frequent starts and stops. It is, moreover, proposed that the system respond only relatively slowly to traffic occupancy variations so that a single car or even a few cars can themselves not have any overriding effect upon the signal timing; instead, the system will respond only to relatively long term changes in traffic conditions.

In FIG. 4, a typical fixed-time controller is shown which comprises a timer and a signal drum. The timer is ordinarily operated synchronously and, at preselected intervals, it produces a pulse of energy on wire 34a which, were it not for the vehicle-actuated control system of this invention, would energize solenoid 30. Solenoid 30 is mechanically coupled to the signal drum so that its actuation causes the signal drum to be advanced to the next phase. Ordinarily, however, front contact 35 is closed so that actuation of solenoid 30 is dependent upon the closing of front contact 36 of relay T2. In addition, the controller is also effective, in the usual manner, to control the signal lamps so that tratlic is permitted to proceed alternately in the two directions of travel.

Assuming that the signal system is a two-phase one, the controller on successive actuations applies energy in turn to wires 31-34 respectively. When wire 31 is energized, the signal displayed for East-West trafiic is green, i.e. phase one green, and relay PGI is also energized. Upon the next cnergization of solenoid 30, the controller is advanced to a position whcrc wire 32 becomes energized. This causes the signals for EastWcst traffic which have heretofore been displaying green to instead display a yellow or amber aspect, and relay All then is energized as well. Ordinarily, the signals, governing the North-South direction of travel still remain red at such times. Upon further energization of solenoid 30, wire 33 becomes energized so that the signals controlling North-South tratiic display a, green aspect, and relay PGZ is then picked up. At such time, the signals for East-West traffic are then controlled to display red aspects. Still later, wire 34 is energized and this results in the display of an amber aspect for North-South traffic and the energization of relay A2.

When the controller has operated to the condition wherein relay P61 is picked up and a green signal is being presented to East-West traffic, front contact 37 of this relay PGl closes, and this connects the summing point S2 through this closed front contact 37, back contact 38 of relay PG2, back contact 39 of relay AP, front contact 40 of relay T1, and through capacitor 41 to the cathode of tube V2. The charging voltage that is applied to capacitor 41 and to the cathode tube V2 is thus dependent upon the amplitude of voltage at the summing point S2. As will subsequently be described, the amplitude of this voltage determines how long a time must elapse before relay T2 can again be energized and thus cause a pulse to be applied to solenoid 30 through from contact 36 of this relay T2. When such interval has elapsed, the controller is then advanced a step so that energy appears on wire 32, and this causes relay A1 to pick up. Repeater relay AP is then energized through from contact 43 of relay A1. Front contact 39 of relay AP then closes, and this connects the cathode of tube V2 through capacitor 41, front contact 40 of relay T1, and front contact 39 of relay AP, to the variable tap of potentiometer 42. The voltage that is then applied through capacitor 41 to the cathode of tube V2 is substantially less than the voltage which was previously available at the summing point 52. For this reason, the system will remain for only a relatively short length of time in the condition wherein relay AP is picked up. In other words, after a relatively brief interval, relay T2 will pick up and will again energize solenoid 30 so that the controller will take one more step.

At this next step of the controller, wire 33 becomes energized and this not only provides a green aspect for the signals controlling the traffic in the North-South direction but also immediately connects the cathode of tube V2 through front contact 40, back contact 39 and front contact 38 to the summing point S1. Thereafter,

the length of time that must elapse before relay T2 can again pick up is dependent upon the amplitude of voltage at the summing point S1. It can thus be seen that the controller remains in the phase one and phase two green positions for time intervals which are dependent entirely upon the measure of traffic congestion, i.e. ccupancy, for the approaches governed by the respective signals. As is customary, the duration of an amber signal is unaffected by traffic and is here set by the position of the variable tap on potentiometer 42.

Since the system is one that is ordinarily continually in operation, it is believed desirable to describe its operation from the moment that a signal change has just occurred. It will be assumed that the signal change just.

occurring is that in which relay A2 is picked up directly after an interval in which relay PG2 was picked up so that, upon the release of relay A2, it will be necessary for relay PGl to be energized (FIG. 5, lines H and J). At the time of a change in phase, tube V3 has always just been rendered nonconductive. This results from the fact that tube V2 becomes conductive as its cathode potential decreases in response to the charging of capacitor 41. When tube V2 conducts, its plate potential abruptly decreases, and this drives the grid-cathode to tube V3 below cutoff. Also, as will subsequently appear, capacitor 45 connected between the control grid of tube V4 and ground will then be fully discharged so that tube V4 will then also be cut off (FIG. 5, lines F and G) since its cathode is maintained substantially above ground by reason of its being connected to a variable tap on potentiometer 47 which is connected between (8+) and ground. Therefore, the cutting off of tube V3 at the inslant of the change in cycle (FIG. 5, line (I), together with the fact that tube V4 will previously have been cut off as well, means that relay Tl included in their common plate circuit will drop away.

One effect of the dropping away of relay T1 is that back contact 40 closes and this discharges capacitor 41 through discharge resistor 48. Another effect is that back contact 49 of relay Tl closes, and this results in the energization of relay T2 through a circuit including this contact, the winding of relay T2, and resistor 50. As soon as relay T2 picks up, its front contact 36 closes,

and this energizes solenoid 30 so as to advance the con-- troller one step. Energy is then removed from wire 34 and applied instead to wire 31 so that relay A2 drops away and relay PGl picks up.

The picking up of relay T2 not only has the effect of advancing the controller in the manner just described but also opens the discharge circuit for capacitor at the now open back contact 51. Until this time, relays T1 and T2 have been respectively picked up and dropped away so that a discharge circuit for capacitor 45 has been closed through resistor 46, back contact 51, and front contact 49, to ground. Now, however, with the opening of this discharge circuit, capacitor 45 charges from (B+) through resistor 52. As shown in FIG. 5, at line F, the charging time is made relatively slow by providing a large value of resistance for resistor 52 so that perhaps one second or so is required before capacitor 45 is charged sufliciently positive to overcome the negative grid-cathode bias and permit tube V4 to conduct. As soon as tube V4 becomes conductive, relay T1 is again energized.

Before relay Tl picks up, however, the advance in step of the controller has resulted in the dropping away of relay A2 and the picking up of relay PGl and this connects the summing point S2 through front contact 37, back contact 38 of relay P62, and back contact 39 of relay AP, as well as front contact 40 of relay T1, to the lefthand terminal of capacitor 41. Tube V2 is immediately driven to cutoff by the application to its cathode of a high positive potential. The exact amplitude of the cathode potential is, of course, dependent upon pavement occupancy on the East-West approaches. The voltage at the summing point will always exceed 100 volts and thus the cathode potential of tube V2 will also exceed 100 volts and this will be more positive than the fixed 100 volt potential applied to the control grid of this tube so that tube V2 will abruptly be driven to cutoff (FIG. 5, lines A and B). When this happens, the plate potential of tube V2 abruptly rises since the flow of current through its plate resistor 53 is then terminated. This rise in plate potential is sufficient to drive the control grid of V3 into the conductive region so that relay Tl can now be maintained picked up by the plate current of this tube and now no longer requires plate current from tube V4 (line C). This picking up of relay T1 will, incidentally, have had the further effect of causing relay T2 to drop away.

With relay T1 picked up and relay T2 dropped away, capacitor 45 is abruptly discharged through front contact 49 of relay T1 and back contact 51 of relay T2. Tube V4 is then cut off so that continued energization of the winding of relay T1 is then dependent upon the conductive status of tube V3.

Throughout the time that the cathode of tube V2 is maintained at an elevated potential because of the voltage applied to it through capacitor 41, a substantially constant charging current flows through the plate-cathode circuit of tube V1. The relatively high cathode resistor 55 ensures that the plate-cathode current of tube Vl can vary only over a quite small range. Thus, capacitor 45 charges substantially linearly through tube V1.

Eventually, however, the cathode potential of tube V2 is reduced to the point where this tube again becomes corttluctivc. The length of time that must elapse before this occurs is dependent upon the amplitude of positive voltage which has been applied through capacitor 41. When tube V2 becomes conductive, its plate voltage is substantially lowered in value and this causes tube V3 to be cut off so that relay T1 drops away. Following this, the operation occurs again in the same sequence, with relay T2 picking up to thereby provide energy to the controller and with capacitor 45 again charging so that tube V4 will eventually become conductive and thereby result in the reenergization of relay T1. This time, however, upon the picking up of relay T1, at different one of the relays P61, P62 and AP will be energized so that a different amplitude of the charging voltage will be applied through capacitor 41 to the cathode of tube V2.

Upon each energization of'relay T2, energy is applied through front contact 36 of relay T2, from contact 35 of the failure timer FT, and front contact 56 of the timer FT, and through its winding to ground. As a result, the timer FT is periodically energized andthus maintains its front contacts 56 and 35 closed. The timer F1" is of the type that'will remain actuated as long as it receives such periodic energization, but is constructed to become actuated to an opposite condition when it receives no energy for a time in excess of some predetermined interval. Therefore, if there should be any failure in the circuit organization just described which effects the periodic picking up of relay T2, the timer FT will operate and will close its back contact 35. Thereafter, the controller will receive its input only from the timer which may be synchronously driven and thus advance the controller at fixed intervals to provide the various signal phases.

Once timer FT is operated because of a failure to energize it throughout some predetermined interval, the opening of its front contact 56 thereafter prevents it from becoming reenergized unless the manually operated pushbutton contact 57 is closed and maintainedclosed until have been available to energize the winding of the failure timer Fl" from the signal timer 60.

There has been disclosed herein a system for the control of traffic signals which is particularly adapted to be used in connection with the conventional type of fixed time controller which is currently in widespread use. It therefore becomes possible to utilize the apparatus of this invention directly in connection with such a fixed time control and when this is done, the added advantage of being able to use the fixed timer in the event of circuit failure constitutes a desirable feature. However, it should be clearly understood that the use of such a fixed timer is not essential to the operation of the invention 'but that the controller can be operated directly and exclusively by the apparatus in this invention, such as by the periodic actuation of a relay corresponding to the relay T2 of FIG. 4.

Having described a traffic actuated signal control system as one specific embodiment of this invention, I desire itto be understood that various modifications, adaptations, and alterations may be made to the specific form shown without in any manner departing from the spirit or scope of this invention.

What I claim is:

1. A system for the control of traffic control devices regulating traflic approaching an intersection of at least two streets comprising in combination, vehicle presence detection means for at least one of said streets defining a detection zone traversed by vehicles approaching said intersection, said detection means generating a signal in response to each vehicle traversing said detection zone which signal has a value proportional to the time said vehicle occupies said detection zone, means responsive to said detection means for generating a traffic control manifestation having at least one characteristic proportional to the ratio of the cumulative values of said signals occurring throughout any given measuring interval to the total duration of said measuring interval, and control means responsive at least in part to said one characterthere has been a change in phase so that energy will Ml istic of said traffic control manifestation for controlling the duration of at least one of the traflic control indications displayed by said traffic control devices at said intersection.

2. The trafiic control system of claim 1 wherein there is at least one vehicle detection means and a corresponding control manifestation generating means for each of at least two different'interfering approaches to said intersection, whereby two separate traffic control manifestations are generated for the respective interfering approaches, said control means being governed by both said traffic control manifestations to control said trafiic control devices to'display proceed indications for said interfering approaches whose durations are respectively proportional to the relative values of said one characteristic of the respective control manifestations.

3;The system as defined in claim 1 wherein said vehicle presence detection means includes at least one detection zone defining element which is so positioned and directed that the spacing between said detection zone and said intersection is sufiiciently greatthat said detection zone is ordinarily not occupied 'by a vehicle which is part of a queue of vehicles waiting to proceed toward andlthrough said intersection.

4. The invention as defined in claim 1 wherein said trafiic control manifestation acts on said control means to adjust the duration of the proceed indication pre sentedto traffic a-pproachingsaid intersection along said one street.

5. The invention-as defined in claim 1 wherein said system'furtherincludes a storage means for storing each said traffic control manifestation, said storage means having a long time constant permitting said manifestation to vary only slowly and not be significantly affected by only a few vehicles.

6. In a system for controlling vehicular traffic on interfering traflic lanes by traffic signals for the respective lanes which are selectively controllable to provide at least stop and proceed indications, the combination comprising, a controller for said signals successively operable through successive conditions on each of successive cycles, said controller in each of selected conditions controlling said signals to display a proceed indication for traffic on a respective one of said interfering lanes while at the same time displaying a stop indication for other interfering lanes, control means for at times advancing said controller from one condition to the next, vehicle presence detector means for each of said interfering lanes defining a vehicle detection zone and producing an electrical signal for each detected vehicle having a duration corresponding substantially to the length of time required for said vehicle to traverse the respective zone, circuit means associated with and controlled by each said vehicle detector means for producing a continuous output signal whose amplitude is substantially proportional to the percentage of time that said electrical signal is produced by said vehicle detector means throughout any given interval, said control means advancing said controller from any selected condition wherein a proceetf signal indication is presented to a particular one of said lanes to its next condition only after a length of time which is variable in accordance with the amplitude of said continuous output signal provided by the particular circuit means associated with said one particular lane.

7. In a system for the control of trafiic control devices at an intersection which accord right-of-way alternately to the respective approaches to said intersection the combination comprising, a vehicle presence detector means for detecting vehicles on at least one of said approaches, said detector means generating a signal in response to each vehicle traversing said detection zone which signal has a duration proportional to the time required for the detected vehicle .to pass a given point, lane occupancy manifestation generating means comprising:

(l) a storage means for said lane occupancy manifestation;

(2) circuit means controlled by said detector for adding to the existing stored manifestation in response to each vehicle detected an amount proportional to the integral over said occupancy time interval of the difference between:

(a) a predetermined constant value in excess of said manifestation; and (b) said manifestation;

(3) second circuit means for subtracting from said manifestation in proportion to the time integral of the existing magnitude of said manifestation over all time intervals other than said occupancy time intervals;

and control means responsive to the existing magnitude of said lane occupancy manifestation for controlling the duration of accord of right-of-way presented by said control devices to at least one of said approaches.

8. The system of claim 7 in which said control means vontrols the duration of the "proceed" indication presented by said trafiic control devices to vehicles on said one approach having said vehicle presence detector in accordance with said existing value of said lane occupancy manifestation.

9. The system of claim 7 wherein a vehicle presence detector and respective lane occupancy manifestation generating means are provided for each of a plurality of said approaches and said control means controls the proceed indications alternately accorded to each of said approaches to have a duration which is dependent upon the magnitude of the manifestation provided by the corresponding occupancy manifestation generating means.

10. In a system for controlling traffic control devices at an intersection of mutually interfering approaches, a vehicle presence detector for at least one of said approaches defining a corresponding detection zone, said detector having output means operated to a vehicle registering condition in response to each detected vehicle for an occupancy time interval which substantially equals the time required for the detected vehicle to traverse said detection zone, means responsive to said detector for generating a signal whose magnitude is substantially proportional to the percentage of time throughout any given interval that said detector output means is in its said vehicle registering condition, and control means for said traffic control devices responsive to said signal for adjusting the duration of at least one traffic governing indication presented by said control devices to one of said approaches.

11. The system of claim 10 in which said system further includes, a storage means for storing said signal, means controlled by said detector for adding to the stored value of said signal throughout each said occupancy time interval at a rate proportional to the difference in magnitudes of the then-existing value of said signal and a predetermined higher reference value, and means for subtracting from the stored signal throughout intervals between successive occupancy time intervals at a rate proportional to the existing magnitude of said signal.

12. The system as defined in claim 11 in which said control means is responsive to said signal generated in response to vehicle detections by said presence detector means for a particular approach and acts on said control means to adjust the proceed" indication accorded to said magnitude of said signal.

13. The system of claim 11 wherein said subtracting means subtracts from the stored signal at a rate proportional to the difference between the existing magnitude of said signal and a predetermined lower reference value.

14. The system of claim 13 wherein adjusting means adjusts the value of at least one of said higher and lower predetermined values to thereby adjust the maximum andl minimum attainable values of said signal respective y.

References Cited by the Examiner UNITED STATES PATENTS 2,288,601 7/1942 Barker 340-37 2,750,576 6/1956 Beaubien 340-37 2,834,001 5/1958 Wilcox 340-37 2,925,583 2/1960 Jeffers 340-37 3,047,838 7/1962 Hendricks 340-35 3,079,587 2/1963 Barker 340-37 3,120,651 2/1964 Hendricks 340-35 NEIL C. READ, Primary Examiner.

THOMAS B. HABECKER, Examiner.

Claims (1)

1. A SYSTEM FOR THE CONTROL OF TRAFFIC CONTROL DEVICES REGULATING TRAFFIC APPROACHING AN INTERSECTION OF AT LEAST TWO STREETS COMPRISING IN COMBINATION, VEHICLE PRESENCE DETECTION MEANS FOR AT LEAST ONE OF SAID STREETS DEFNINING A DETECTION ZONE TRAVERSEDB BY VEHICLES APPROACHING SAID INTERSECTION, SAID DETECTION MEANS GENERATING A SIGNAL IN RESPONSE TO EACH VEHICLE TRAVERSING SAID DETECTION ZONE WHICH SIGNAL HAS A VALUE PROPORTIONALL TO THE TIME SAID VEHICLE OCCUPIES SAID DETECTION ZONE, MEANS RESPONSIVE TO SAID DETECTION MEANS FOR GENERATING A TRAFFIC CONTROL MANIFESTATION HAVING AT LEAST ONE CHARACTEISTIC PROPORTIONAL TO THE RATIO OF THE CUMULATIVE VALUES OF SIGNALS OCCURRING THROUGHOUT ANY GIVEN MEASURING INTERVAL TO THE TOTAL DURATION OF SAID MEASURING INTERVAL AND CONTROL MEANS RESPONSIVE AT LEAST IN PART TO SAID ONE CHARACTERISTIC OF SAID TRAFFIC CONTROL MANIFSTATION FOR CONTROLLING THE DURATION OF AT LEAST ON OF THE TRAFFIC CONTROL INDICATIONS DISPLAYED BY SAID TRAFFIC CONTROL DEVICES AT SAID INTERSECTION.

Images