JP5255791B2 - Traffic signal control apparatus and traffic signal control method - Google Patents

Description

  The present invention relates to an autonomous distributed traffic signal control device and a traffic signal control method in an autonomous distributed traffic signal control device.

In an autonomous decentralized traffic signal control system, each traffic signal control device installed at each intersection performs traffic signal control at its own intersection, and predicts traffic demand at its own intersection in the near future, for example, up to a few minutes ahead. Then, signal control is performed according to the predicted traffic demand. Specifically, by predicting outflow traffic flow data to / from other intersection control devices, it is possible to predict time-series changes in arrival traffic flow at the intersection and to optimize signal control at the intersection. The parameter is determined and signal control is performed (for example, refer to Patent Document 1).
JP 2005-182219 A

  In the conventional traffic signal control shown in Patent Document 1 described above, since the signal control parameter is changed based on the predicted traffic demand, there may be a case where appropriate control for excessive traffic demand is not performed. That is, there is a case where appropriate signal control is not performed when there is a traffic jam as a result of a large traffic demand. During traffic jams (supersaturated state), the traffic volume reaches a peak and does not change (does not increase) even if the traffic jams worsen. For this reason, the situation where the predicted traffic volume and the actual demand deviate and traffic congestion is further increased may occur. In addition, when the traffic is congested, the vehicle speed between the intersections becomes slow, so that the arrival timing prediction error becomes large. This was also the cause of the large difference between the predicted traffic volume and the actual traffic demand.

  The present invention has been made in view of the above circumstances, and aims to determine the occurrence of traffic jams in autonomous distributed traffic signal control and realize appropriate signal control according to the determined traffic jam occurrences. Yes.

The first invention for solving the above-described problems is
Communication means (for example, the communication control unit 200 in FIG. 8) that mutually transmits and receives outflow traffic flow information to and from other traffic signal control devices at other intersections, and arrival traffic flow prediction that predicts arrival traffic flows at the own intersection Means (for example, arrival traffic flow prediction unit 110 in FIG. 8) and the number of outflows in each inflow path at the own intersection and the number of outflows by outflow direction from each inflow path to the other road to predict outflow traffic from the own intersection Outflow traffic flow prediction means for predicting the flow (for example, the outflow traffic flow prediction unit 140 in FIG. 8), and signal control means for controlling the signal at the own intersection by changing the signal control parameter of the own intersection based on the prediction result ( For example, an autonomous distributed traffic signal control device (for example, the traffic signal control device 20 of FIG. 8) provided with the control parameter changing unit 130 and the signal control unit 150) of FIG.
Based on the detection result of the vehicle detector disposed in the inflow path of the own intersection, a traffic congestion determination means for determining the occurrence of traffic congestion in the inflow path (for example, the traffic congestion generation determination unit 120 in FIG. 8),
In response to the determination of the traffic jam occurrence determination means, the signal control parameter at the own intersection by the signal control means is changed to a predetermined traffic jam parameter (for example, the control parameter changing unit in FIG. 8). 130)
Is a traffic signal control device further provided.

In addition, the seventh invention,
Send and receive information on the outflow traffic flow to and from other traffic signal control devices at other intersections, predict the arrival traffic flow at the own intersection, and the number of each inflow path at the own intersection and from each inflow path to the other road Autonomous decentralized traffic that predicts the outflow traffic flow from the own intersection by predicting the number of outflows by the outflow direction of the vehicle, and controls the signal of the own intersection by varying the signal control parameter of the own intersection based on the prediction result A traffic signal control method in a signal control device,
Based on the detection result of the vehicle detector disposed in the inflow path of the own intersection, the occurrence of traffic congestion in the inflow path is determined (for example, steps C9 to C15 in FIG. 27),
In response to the determination of occurrence of the traffic jam, the signal control parameter based on the prediction result is changed to a predetermined traffic jam parameter (for example, steps B5 to B7 in FIG. 26).
This is a traffic signal control method.

  According to the first or seventh aspect of the present invention, in the autonomous distributed traffic signal control apparatus, occurrence of traffic congestion in the inflow path is determined based on the detection result of the vehicle detector disposed in the inflow path of the own intersection. In response to the determination of the occurrence of the traffic jam, the signal control parameter is changed to a predetermined traffic jam parameter. As a result, when there is no traffic jam, the signal control parameter is varied based on the prediction result of the arrival traffic flow, signal control is performed, and the occurrence of traffic jam is determined. Appropriate signal control capable of quickly resolving the generated traffic jam is realized, for example, the signal control is performed by changing to the traffic jam parameter.

The second invention is the traffic signal control device of the first invention,
The congestion occurrence determination means determines the occurrence of congestion for each inflow path based on the detection result of each vehicle sensor disposed in the inflow path of the own intersection (for example, steps C9 to C15 in FIG. 27),
The signal control means at the time of traffic jam is a combination of the presence / absence of occurrence of traffic jam by the inflow path by the traffic jam judgment means from among predetermined parameters for traffic jam for each combination of traffic jam occurrence by inflow path And the traffic control parameter corresponding to is changed to the selected traffic control parameter (for example, steps B5 to B7 in FIG. 26).
It is a traffic signal control device.

  According to the second aspect of the present invention, the occurrence of traffic congestion for each inflow path is determined based on the detection result of the vehicle detector, and the traffic congestion time parameter determined in advance for each combination of presence or absence of traffic congestion for each inflow path. The signal control parameter of the own intersection is changed to the traffic-time parameter corresponding to the determined combination of the occurrence or non-occurrence of traffic for each inflow path. Thus, for example, depending on the combination of the inflow paths determined to have traffic jams, the inflow path to which inflow path is relatively changed for each inflow path, for example, the display time for displaying the right of passage is relatively changed. More detailed traffic signal control is realized depending on whether or not the problem occurs.

The third invention is
Communication means for transmitting and receiving outflow traffic information to and from other traffic signal control devices at other intersections, arrival traffic flow prediction means for predicting arrival traffic flows at the own intersection, and retention of each inflow path at the own intersection Outflow traffic flow prediction means to predict the outflow traffic flow from the own intersection by predicting the number of outflows by number of outflows from each inflow path to the other road and the signal control parameters at the own intersection based on the prediction result An autonomous decentralized traffic signal control device equipped with a signal control means for controlling the signal at its own intersection,
Based on the detection result of the vehicle detector disposed in the inflow path of the own intersection, further comprising a congestion occurrence determination means for determining the occurrence of traffic congestion in the inflow path,
The signal control means has a correction means for correcting the signal control parameter based on the determination result when the determination by the congestion occurrence determination means is made.
It is a traffic signal control device.

Further, the eighth invention is
Send and receive information on the outflow traffic flow to and from other traffic signal control devices at other intersections, predict the arrival traffic flow at the intersection, and the number of stays at each inflow path at the own intersection and from each inflow path to the other road Autonomous decentralized traffic signal that predicts outflow traffic flow from its own intersection by predicting the number of outflows by outflow direction, and controls the signal at its own intersection by changing the signal control parameter of its own intersection based on the prediction result A traffic signal control method in a control device,
Based on the detection result of the vehicle detector arranged in the inflow path of the own intersection, determine the occurrence of traffic congestion in the inflow path,
Correcting the signal control parameter based on the determination result when the occurrence of the traffic jam is determined,
This is a traffic signal control method.

  According to the third or eighth aspect of the present invention, in the autonomous distributed traffic signal control apparatus, occurrence of traffic congestion in the inflow path is determined based on the detection result of the vehicle detector disposed in the inflow path of the own intersection. When the occurrence of the traffic jam is determined, the signal control parameter is corrected based on the determination result. As a result, when there is no traffic jam, the signal control parameter calculated based on the prediction result of the outflow traffic flow is varied to perform signal control, and when the occurrence of traffic jam is determined, the calculated signal control parameter is Appropriate signal control capable of quickly eliminating the generated traffic jam, such as signal control with correction to an optimum value for cancellation, is realized.

The fourth invention is the traffic signal control device of the third invention,
The correction means is configured to display the former indication when there is an indication that gives the right of passage to the inflow route where the occurrence of the traffic jam is determined and an indication that gives the right of passage to the inflow route where the occurrence of the traffic jam is not determined. The traffic signal control device corrects the signal control parameter so that the indicated time is longer than when the determination by the congestion occurrence determination means is not made.

  According to the fourth aspect of the invention, the correction of the signal control parameter is such that the display time for giving the right of passage to the inflow path where the occurrence of the traffic jam is determined is longer than when the traffic jam is not judged. Done. As a result, the length of time that the right of passage is on the inflow path where the occurrence of the traffic jam has been determined becomes longer, so that the traffic jam in the inflow path is eliminated and more appropriate signal control is realized.

A fifth invention is the traffic signal control device according to any one of the first to fourth inventions,
Further comprising a traffic jam occurrence estimating means for estimating the occurrence of traffic jam for each inflow path based on the number of staying in each inflow path of the own intersection predicted by the outflow traffic flow prediction means,
The occurrence of traffic jam is determined based on a detection result of the vehicle detector and an estimation result of the traffic jam generation estimation unit,
It is a traffic signal control device.

  According to the fifth aspect of the present invention, the occurrence of traffic congestion for each inflow path is estimated based on the predicted number of stays in each inflow path at its own intersection, and based on the detection result of the vehicle detector and the estimation result of the occurrence of congestion. The occurrence of traffic congestion is determined. In other words, in addition to the detection result of the vehicle detector, it is possible to determine the occurrence of traffic congestion based on the predicted number of stays, so that more accurate determination of traffic congestion is possible. Signal control is realized.

The sixth invention is the traffic signal control device of the fifth invention,
The congestion occurrence determination means determines the occurrence of congestion for each inflow path based on the detection result of the vehicle sensor for the inflow path where the vehicle sensor is disposed, and the vehicle sensor is disposed. The inflow route that is not determined based on the estimation result of the inflow route by the congestion occurrence estimation means,
It is a traffic signal control device.

  According to the sixth aspect of the invention, the occurrence of traffic congestion for each inflow path is determined based on the detection result of the vehicle detector in principle, and is determined based on the predicted number of staying when the vehicle is not installed. In this way, by using the determination result based on the predicted staying number as an auxiliary, appropriate signal control in accordance with the actual traffic situation is realized.

  According to the present invention, in the autonomous decentralized traffic signal control device, the occurrence of traffic jam in the inflow path is determined based on the detection result of the vehicle detector disposed in the inflow path of the own intersection, and the determination of the occurrence of this traffic jam. Accordingly, the signal control parameter is changed to a predetermined traffic jam parameter. As a result, when there is no traffic jam, the signal control parameter is varied based on the prediction result of the arrival traffic flow, signal control is performed, and the occurrence of traffic jam is determined. Appropriate signal control capable of quickly resolving the generated traffic jam is realized, for example, the signal control is performed by changing to the traffic jam parameter.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

[System configuration]
FIG. 1 is an overall configuration diagram of an autonomous distributed traffic signal control system 1 of the present embodiment. According to the figure, an autonomous decentralized traffic signal control system 1 includes a central management device 10 installed at a control center and a plurality of traffic signal control devices 20 installed at each intersection via a transmission line N. Connected and configured. The traffic signal control device 20 is connected to a plurality of traffic signals 30 and a vehicle detector 40 provided at an intersection where the device is installed (own intersection). The traffic signal control device 20 transmits / receives data to / from other traffic signal control devices 20 such as adjacent intersections, and each traffic signal device provided at the own intersection based on a detection result signal or the like by the vehicle detector 40. 30 is controlled.

  FIG. 2 is a diagram illustrating an example of an intersection to be controlled by the traffic signal control device 20. According to the figure, the intersection is a cross intersection having four inflow paths A to D, and a traffic signal 30 and a vehicle detector 40 are installed in each of the inflow paths A to D. The figure is an example of an intersection, and is not limited to this intersection, but can be applied to intersections of other shapes such as a three-way intersection or a T-shaped intersection.

  The vehicle detector 40 is, for example, an ultrasonic type or an optical type, and detects the presence of a vehicle passing through the installed position. Then, as a sensing result, a time occupancy ratio that is a ratio of the time when the presence of the vehicle is sensed with respect to a certain time is output. The vehicle sensor 40 may be an image type. In this case, the traffic volume, the speed of the vehicle, and the queue length (congestion length) are measured by imaging the vehicle traveling in each lane and performing image processing.

  FIG. 3 is a diagram illustrating an installation example of the vehicle detector 40. According to the figure, a point (determination point) that is a target for determining whether or not a traffic jam has occurred is defined in each inflow path at the intersection. At some of these determination points, vehicle detectors 40 that are targeted for detection are installed. In the figure, determination points A-1 and A-2 of the inflow channel A, determination point B-1 of the inflow channel B, determination point C-1 of the inflow channel C, and determination point D-1 of the inflow channel D are respectively shown. A vehicle sensor 40 is installed.

  FIG. 4 is a diagram showing an example of a method of displaying the intersection. In the figure, a solid line arrow indicates a flow line of a vehicle to which a right of passage is given. That is, at this intersection, a current indication 1φ that gives the right of passage to the vehicle traffic on the inflow paths A and C and a current indication 2φ that gives the right of passage to the vehicle traffic on the inflow routes B and D are alternately displayed. This figure is an example of the presenting method, and the present invention is not limited to the two presenting methods, and a multiple presenting method having three or more presenting methods may be used.

[Outline of signal control]
FIG. 5 is a schematic diagram of signal control in the traffic signal control device 20. The traffic signal control device 20 first predicts the traffic flow arriving at its own intersection based on (1) the predicted outflow traffic flow information received from the traffic signal control device 20 at the adjacent intersection. Next, (2) based on the predicted arrival traffic flow (predicted arrival traffic flow), the control parameters (cycle length C and split) of each traffic signal 30 at its own intersection are changed. Then, (3) according to the changed control parameter, each traffic signal 30 at the own intersection is controlled. Further, (4) based on the predicted arrival traffic flow and the signal control parameter, the traffic flow flowing out from the own intersection (outflow traffic flow) is predicted. Then, the traffic flow information predicted to flow out (predicted outflow traffic flow information) is transmitted to the traffic signal control device 20 at each adjacent intersection.

[principle]
(1) Runoff prediction The outflow traffic flow prediction principle will be explained. 6 and 7 are diagrams illustrating the principle of prediction of the outflow traffic flow, and show the inflow path A or C in FIG. As shown in FIG. 2, the inflow paths A and C have a two-lane straight turn left lane and a one-lane right turn lane. Further, the saturated traffic flow is “0.5 [unit / s]” equally in any lane.

  A vehicle that has arrived at the inflow path travels straight, turns left or right in the intersection, and flows out from the inflow path (direction) corresponding to the travel direction. In which direction the vehicle that has arrived on the inflow path travels is determined stochastically. Specifically, the probability of going straight ahead (straight forward rate) is “0.7 (70%)”, the probability of going left turn (left turn rate) is “0.1 (10%)”, and turns right The probability of proceeding (right turn rate) is “0.2 (20%)”.

(1-1) When there is no right of traffic FIG. 6 is an explanatory diagram of the outflow prediction when there is no right of traffic on the inflow channel. In the figure, in order from the left side in the figure, the arrival traffic flow of the inflow path, the planned indication predicted from the signal control parameters, and the number of staying in each lane are shown as the time axis t in the lower direction in the figure. When there is no right to pass, the vehicle that has arrived at the inflow path stays in the inflow path as it is. That is, all the vehicles that have arrived at the inflow path at time t when the scheduled display is “red” stay in the inflow path. More specifically, of the number of arrivals, the number of vehicles having a ratio of “0.8 (= straight ahead rate“ 0.7 ”+ left turn rate“ 0.1 ”)” arrives at the straight turn left lane, and “0.2 (= The right turn rate)) will arrive in the right turn lane. The number of staying lanes at time t is the number obtained by adding the number of arrivals to the number of staying lanes at the immediately preceding time t.

For example, the number of arrivals at time t ₁ when the scheduled display is “red” is “0.5 [unit]”. That is, “0.4 (= 0.5 × 0.8) [unit]” arrives at the straight turn left lane, and “0.1 (= 0.5 × 0.2) [unit]” enters the right turn lane. arrive. Then, the number of staying in each lane at time t ₁ is obtained by adding the number of arrivals “0.4 [units]” to the number of staying “3.2 [units]” at the immediately preceding time t ₀ for the straight turn left lane. 3.6 [units] ”, and for the right turn lane,“ 0.3 [units] is obtained by adding the arrival number “0.1 [units]” to the number of remaining units “0.2 [units]” at the previous time t ₀ . ] ”.

(1-2) When there is a right of traffic FIG. 7 is an explanatory diagram of the outflow prediction when there is a right of traffic in the inflow channel. In the figure, in order from the left side in the figure, the arrival traffic flow of the inflow path, the planned display, the number of staying in each lane, and the outflow traffic flow are shown as time t in the downward direction in the figure. When there is a right of passage, the vehicle that has arrived at the inflow path travels along the intersection together with the vehicle that has stayed in the inflow path, and flows out from the other inflow path. That is, at time t when the scheduled display is “blue”, the vehicle that has arrived on the inflow route is “0.8 (= straight ahead rate“ 0.7 ”), similarly to the case where there is no right of traffic shown in FIG. + Left turn rate “0.1”) ”arrives at the straight turn left lane, and“ 0.2 (right turn rate) ”arrives at the right turn lane. Here, the vehicle arriving in each lane is considered to be temporarily staying in the lane together with the vehicle staying in the lane, and the number obtained by adding this arrival number to the staying number at the immediately preceding time t is This is called “temporary residence”.

  Then, the vehicle temporarily staying in each lane travels in a straight direction, a left turn or a right turn direction, and flows out from another inflow path. That is, from the straight ahead left turn lane, the temporarily staying vehicle flows out in the straight direction or the left turn direction. Specifically, from the straight turn left turn lane, the number of vehicles staying at a ratio of “0.7 (straight forward rate) /0.8 (straight forward rate“ 0.7 ”+ left turn rate“ 0.1 ”)” is straight ahead. The number of vehicles with a ratio of “0.1 (left turn rate) /0.8 (straight ahead rate“ 0.7 ”+ left turn rate“ 0.1 ”)” flows out in the left turn direction. Moreover, all the vehicles that have stayed temporarily flow out from the right turn lane in the right turn direction.

  However, the number of outflows from each lane is determined so as not to exceed the saturation traffic flow of the lane. In other words, if the number of temporarily staying is less than or equal to the saturated traffic flow, the number of temporarily staying is the number of outflows from the lane, and if the number of temporarily staying exceeds the saturation traffic flow, the number of saturated traffic flows from the lane. The number of outflows. In the example of FIG. 2, the straight turn left lane has two lanes. Therefore, the saturated traffic flow “0.5 [vehicle / s] per lane” is multiplied by the number of lanes “2” of the lane “1”. .0 [unit / s] ".

For example, the arrival number at time t ₃ will present shows is "blue" is "0.5 [units]." That is, among the arriving vehicles, “0.4 (= 0.5 × (straight ahead rate“ 0.7 ”+ left turn rate“ 0.1 ”))” [car] ”arrives in the straight left turn lane, and“ 0 .1 (= 0.5 × Right turn rate “0.2”) [unit] ”arrives on the right turn lane. Then, in a straight left turn lane, the sum of the arrival number "0.4 [units]" to the residence number at time t ₂ of the previous "3.6 [units],""4.0[units]" is a temporary residence number However, since the temporarily staying number “4.0 [units]” exceeds the saturated traffic flow “1.0 [units / s] (= 0.5 × 2 [lanes])”, it flows out from the straight turn left lane. The number is “1.0 [unit]”. Then, out of the number of outflows “1.0 [unit]”, “0.9 (≈1.0 × 0.7 / 0.8) [unit]” proceeds in the straight direction and flows out. .1 (≈1.0 × 0.1 / 0.8) [unit] ”proceeds in the left turn direction and flows out. Therefore, the residence number at time _{t 3} in the straight left turn lane is "3.0 (= 4.0-1.0) [units]."

On the other hand, in the right turn lane, “0.4 [unit]”, which is obtained by adding the arrival number “0.1 [unit]” to the staying number “0.3 [unit]” at the time t ₂ immediately before, becomes the temporary residence number. However, since the temporary residence number “0.4 [unit]” does not reach the saturated traffic flow “0.5 [unit / s]”, the number of outflows from the right turn lane is “0.4 [unit]”. Become. And all of this outflow number “0.4 [unit]” flows out in the right turn direction. Therefore, the residence number at time _{t 3} in the right turn lane is "0.0 [units]."

  However, in consideration of the time Δt required for the vehicle to pass through the intersection, the vehicle that has flowed out from each lane of the inflow path at time t is the inflow path (route) in the traveling direction at the time (t + Δt) after the intersection passage time Δt. ).

(2) Calculation of Signal Control Parameter Next, the calculation principle of the signal control parameter will be described.

(2-1) Traffic congestion occurrence First, the presence / absence of traffic congestion is determined. Specifically, “detection” of the occurrence of traffic jam based on the detection result signal from the vehicle detector 40 and “estimation” of the occurrence of traffic jam based on the predicted arrival traffic flow are performed. The presence / absence of traffic jam is determined based on the estimation result.

  As “detection” of the presence or absence of occurrence of traffic congestion, the presence or absence of occurrence of congestion in each inflow path is determined based on the detection result signal from the vehicle detector 40. Specifically, for each determination point of each inflow path, a past predetermined period (for example, 5 minutes) based on the detection result signal input from the vehicle detector 40 installed at the determination point. An average value of the occupation ratio is calculated, and whether or not a traffic jam has occurred is detected based on whether or not the calculated average occupation ratio exceeds a predetermined threshold (for example, 0.4). That is, if the average occupancy exceeds the threshold, it is detected that there is a traffic jam “present”, and if it does not exceed it, it is detected that there is no traffic jam. At this time, the average occupancy rate is calculated using only the occupancy rate in the time zone in which the right of passage is on the inflow path of the determination point. However, at the judgment point where the vehicle sensor 40 is not installed, or at the judgment point where the vehicle sensor 40 is installed but the sensing result signal is not received due to a failure or the like, the occurrence of traffic congestion by the vehicle sensor 40 is caused. The detection result is “unknown”.

  When the vehicle sensor 40 is realized by an image type vehicle sensor, the presence or absence of occurrence of traffic jam may be detected based on the measured queue length (traffic jam length). Specifically, whether or not a traffic jam has occurred at the determination point is detected according to whether or not the queue length has reached each determination point.

  Further, as the “estimation” of the occurrence of traffic congestion, the presence / absence of occurrence of congestion in each inflow path is determined based on the staying number calculated in the process of predicting the outflow traffic flow. Specifically, for each inflow path, the next timing of switching from the present signal control parameter to the present display that has the right of passage in the inflow path is determined. Next, the number of stays immediately before the determined switching timing is determined. The number of staying is calculated in the process of predicting the outflow traffic flow described above. Then, based on the determined staying number, the predicted traffic jam length (predicted traffic jam length) is calculated. The predicted traffic jam length is calculated, for example, by multiplying the determined number of stays by the total length (for example, 4 [m]) per vehicle including the inter-vehicle distance.

  Subsequently, for each determination point of each inflow path, whether or not a traffic jam has occurred is estimated depending on whether or not the predicted congestion length of the inflow path has reached the determination point. That is, if the predicted traffic jam length has reached the determination point, it is estimated that the traffic jam has occurred, and if it has not reached, the traffic jam is estimated to be “absent”.

  Then, based on these detection results and estimation results, it is determined whether or not there is a traffic jam at each determination point. That is, it is determined that there is a traffic jam “existing” for a determination point with a detection result “present”, and an estimation result is directly used as a determination result for a determination point with a detection result “not present” or “unknown”.

(2-2) When traffic jam occurs “Yes” When a traffic jam occurs and is determined to be “Yes” at least at one determination point, a predetermined traffic jam occurs. Change to the value of the control parameter for the hour. That is, the control parameter value for the occurrence of traffic jam is determined in advance for each combination of occurrence of traffic jam in each inflow path, and the signal associated with the determined combination of occurrence of traffic jam Change to the value of the control parameter.

式（１）において、Ｌは交差点の損失時間である。λは交差点の需要率であり、各現示ｉの需要ρｉの合計値である。この交差点需要率λは、例えば図４に示す現示方式の交差点では、現示１φ，２φそれぞれの需要率ρ１，ρ２の合計値となる。 (2-3) When traffic jam occurs “None” On the other hand, when it is determined that traffic jam occurs “None” at all judgment points as a result of the judgment of traffic jam occurrence, Based on this, signal control parameters are calculated and changed. That is, the cycle length C is calculated according to the equation (1).

In equation (1), L is the loss time at the intersection. λ is the demand rate at the intersection, and is the total value of the demands ρi of each display i. This intersection demand rate λ is the total value of the demand rates ρ1 and ρ2 of the current indications 1φ and 2φ, for example, at the intersection of the current method shown in FIG.

  Next, the split is calculated. The split is calculated so that the time (C−L) obtained by subtracting the loss time L from the cycle length C is distributed in proportion to the demand rate ρi of each indication i.

[Traffic signal control device]
FIG. 8 is a block diagram showing an internal configuration of the traffic signal control device 20. According to the figure, the traffic signal control device 20 includes a processing unit 100, a communication control unit 200, and a storage unit 300.

  The processing unit 100 is a program or data stored in the storage unit 300, or data received from an external device (mainly, traffic signal control device 20 at an adjacent intersection) via the communication control unit 200 (predicted outflow traffic flow information). ) Etc., various processes such as overall control of the traffic signal control device 20 are performed. The processing unit 100 is realized by a CPU, for example. In addition, the processing unit 100 includes an arrival traffic flow prediction unit 110, a congestion occurrence determination unit 120, a control parameter change unit 130, an outflow traffic flow prediction unit 140, and a signal control unit 150.

  Note that the configuration of an intersection (self-intersection) to be controlled by the traffic signal control device 20 is defined by the self-intersection configuration table 321. FIG. 9 shows an example of the data configuration of the own intersection configuration table 321. According to the figure, the self-intersection configuration table 321 stores, for each inflow path 321a at the self-intersection, a lane 321b constituting the inflow path and the number of lanes 321c are associated with each other.

  Further, the determination points set at the own intersection are defined in the traffic congestion determination point table 322. FIG. 10 shows an example of the data configuration of the traffic jam determination point table 322. According to the figure, the traffic congestion determination point table 322 associates a predetermined determination point 322b, its position 322c, and whether or not the vehicle detector 40 is installed 322d for each inflow path 322a of its own intersection. Storing. The position 322c is represented by the distance from the stop line.

  The arrival traffic flow prediction unit 110 calculates a traffic flow predicted to arrive (predicted traffic flow) based on the predicted outflow traffic flow data 331 at the adjacent intersection received from the traffic signal control device 20 at the adjacent intersection. Specifically, based on the predicted outflow traffic flow data 331 of the adjacent intersection, the outflow traffic flow from the inflow path connected to the inflow path of the own intersection among the inflow paths of the adjacent intersection from the adjacent intersection to the own intersection The travel time will be delayed and the traffic will arrive at the intersection. The arrival traffic flow prediction unit 110 repeatedly executes the calculation of the predicted arrival traffic flow at predetermined time intervals.

  The predicted outflow traffic flow data 331 is data of a traffic flow (predicted outflow traffic flow) predicted to flow out from the intersection. FIG. 11 shows an example of the predicted outflow traffic flow data 331. According to FIG. 6A, the predicted outflow traffic flow data 331 includes predicted outflow traffic flow data 331A at its own intersection and predicted outflow traffic flow data 331B at each adjacent intersection. The predicted outflow traffic flow data 331A of the own intersection is generated by the outflow traffic flow prediction unit 140, and the predicted outflow traffic flow data 331B of the adjacent intersection is data received from another traffic signal control device 20.

According to FIG. 5B, the predicted outflow traffic flow data 331 stores the number of outflows 331b from each inflow path of the corresponding intersection in association with each time 331a within the prediction target time range. Time 331a is the time range of the prediction target, after the time t ₀ for a predetermined time (for example, 200 seconds) a predetermined time interval up to time t _n (for example, 1 second interval) it is time to continuous. The start time t ₀ of the estimated time range is the current time at which the relevant outflow prediction is performed.

The predicted arrival traffic flow data 332 is data of traffic flow (arrival traffic flow) predicted to arrive at the own intersection. FIG. 12 shows an example of the data configuration of the predicted arrival traffic flow data 332. According to the figure, the predicted arrival traffic flow data 332 stores the number of arrivals 332b to each inflow path at the own intersection in association with each time 332a within the time range to be predicted. Time 332a, similar to the predicted outflow traffic flow data 331, from the time _{t 0} is a time range prediction target until time _{t n} after a predetermined time (e.g., after 200 seconds), a predetermined time interval (e.g., 1 second (Interval).

  The traffic jam occurrence determination unit 120 determines whether or not a traffic jam has occurred. Specifically, first, based on the detection result signal input from the vehicle detector 40, the presence / absence of occurrence of traffic congestion in each inflow path at the own intersection is detected. From the vehicle detector 40, the value of the time occupancy rate at the corresponding determination point is output as a detection result signal, and the detection result signal from the vehicle detector 40 is stored as the detection result data 341.

FIG. 13 shows an example of the data configuration of the sensing result data 341. According to the figure, the sensing result data 341 stores a current indication 341b and a sensing result 341c at each determination point in association with each time 341a. Time 341a is before the present time _{t 0} for a predetermined time (for example, 100 seconds before) up to time _{t -100} of a successive times at predetermined time intervals (e.g., intervals of one second). The current indication 341b is an actual current indication. The sensing result 341c stores the time occupation rate acquired from the vehicle detector 40 installed at each determination point. The sensing result data 341 is updated to the latest value at a predetermined time (for example, each one second).

  The traffic congestion determination unit 120 refers to the sensing result data 341 and calculates an average (average occupancy rate) of the time occupancy rate during the period when the right of passage is given to the corresponding inflow path for each determination point. For example, at the intersection shown in FIG. 3, for the determination point A-1, the average value of the time occupancy rate during the period of the current 1φ where the right of passage is given to the corresponding inflow path A is calculated. If the calculated average occupancy exceeds a predetermined threshold (for example, “0.2”), it is assumed that the occurrence of traffic jam “present” is detected at the determination point. Is detected.

  The result of detecting whether or not a traffic jam has occurred by the traffic jam occurrence determination unit 120 is stored in the traffic jam detection result data 342. FIG. 14 shows an example of the data configuration of the traffic jam detection result data 342. According to the figure, the congestion detection result data 342 stores an average occupation ratio 342b and a detection result 342c in association with each determination point 342a. The detection result 342c stores a value indicating occurrence of traffic jam “present”, “absent”, or “unknown”.

  Further, the traffic jam occurrence determination unit 120 estimates whether or not a traffic jam in the near future occurs in each inflow path at the own intersection based on the outflow traffic flow predicted by the outflow traffic flow prediction unit 140. In other words, for each inflow path, the next timing of switching from the current signal control parameter to the display that does not have the right to pass through the inflow path is determined. Next, with reference to the staying number data 334, the staying number predicted in the inflow path immediately before the determined switching timing is determined. Subsequently, the predicted congestion length is calculated by multiplying the calculated staying number by the predetermined total length of one vehicle. Then, for each inflow path, by comparing the calculated predicted congestion length with the position of each determination point (distance from the stop line), the presence / absence of congestion at each determination point is estimated. That is, if the predicted traffic jam length has reached the determination point, it is estimated that there is traffic jam at the determination point, and if it has not reached, the traffic jam is estimated as “no”.

  Here, the current signal control parameter is stored in the signal control parameter data 336. FIG. 15 shows an example of the signal control parameter data 336. According to the figure, the signal control parameter data 336 stores a cycle length 336a and a split 336b. The signal control parameter data 336 is generated / updated by the control parameter changing unit 130 as described later.

  Further, the staying number data 334 is data on the number of staying predicted in each inflow path of the own intersection, and is generated and updated by the outflow traffic flow prediction unit 140 as described later. FIG. 16 shows an example of the data configuration of the staying unit data 334. According to the figure, the staying number data 334 stores the scheduled display 334b and the staying number 334c of each lane in association with each time 334a within the prediction target time range. The scheduled display 334b is a display predicted from the signal control parameter.

  Then, the estimation result of whether or not a traffic jam has occurred by the jam occurrence determination unit 120 is stored in the traffic jam estimation result data 343. In FIG. 17, an example of a data structure of the traffic jam estimation result data 343 is shown. According to the figure, the traffic jam estimation result data 343 stores, for each inflow channel 343a, a predicted staying number 343b, a predicted traffic jam length 343c, a determination point 343d, and an estimation result 343e in association with each other. In the estimation result 343e, a value indicating occurrence of traffic jam “present” or “absent” is stored.

  Thereafter, the traffic jam generation determination unit 120 determines whether or not there is traffic jam on each inflow path based on the detection result and estimation result of the traffic jam occurrence. That is, for each inflow path, the presence / absence of traffic congestion is determined according to the congestion occurrence presence / absence determination table 327 based on the combination of the detection result and the estimation result.

  The congestion occurrence presence / absence determination table 327 is a data table in which a correspondence relationship between a detection result, an estimation result, and a determination result is defined. FIG. 18 shows an example of the data structure of the congestion occurrence presence / absence determination table 327. According to the figure, the congestion occurrence presence / absence determination table 327 stores a detection / estimation result combination 327a and a congestion occurrence determination result 327b in association with each other.

  The determination result of whether or not a traffic jam has occurred by the traffic jam occurrence determination unit 120 is stored in the traffic jam determination result data 344. FIG. 19 shows an example of the data configuration of the traffic jam determination result data 344. According to the figure, the congestion determination result data 344 stores a detection result 344b, an estimation result 344c, and a determination result 344c in association with each determination point 344a.

  The traffic jam generation determination unit 120 repeatedly performs the determination of whether or not a traffic jam has occurred at predetermined time intervals. Therefore, each of the traffic jam detection result data 342, the traffic jam estimation result data 343, and the traffic jam determination result data 344 is updated at predetermined time intervals.

  The control parameter changing unit 130 changes the control parameter of the traffic signal 30 based on the determination result of the occurrence of traffic congestion by the congestion occurrence determining unit 120. Specifically, with reference to the traffic jam determination result data 344, the judgment result of the occurrence of traffic jam at each judgment point is judged. As a result, if it is determined that the occurrence of traffic congestion is “present” even at one location, a combination of occurrence / non-occurrence of traffic congestion at each determination point is determined, and a control pattern corresponding to the determined combination is determined according to the congestion occurrence / non-occurrence combination table 325. Select. Then, referring to the control parameter table 326 at the time of traffic jam, the signal control parameter is changed to the value of the signal control parameter determined as the selected control pattern.

  The congestion occurrence presence / absence combination table 325 is a data table that defines a correspondence relationship between a control pattern and a combination of occurrence / non-occurrence of congestion at each determination point. FIG. 20 shows an example of the data configuration of the jam occurrence presence / absence combination table 325. According to the figure, the jam occurrence presence / absence combination table 325 stores a control pattern 325b in association with each jam occurrence occurrence combination 325a at each determination point.

  The traffic parameter control parameter table 326 is a data table that defines signal control parameters when it is determined that there is a traffic jam “present” (when there is a traffic jam). FIG. 21 shows an example of the data structure of the control parameter table 326 for traffic jam. According to the figure, the control parameter table for traffic jam 326 stores the values of the cycle length 326b and the split 326c that are signal control parameters in association with each control pattern 326a.

  On the other hand, if it is determined that there is no traffic jam at all the determination points, the control parameter changing unit 130 is based on the predicted arrival traffic flow data 332 generated by the arrival traffic flow prediction unit 110. The control parameters (cycle length C and split) of each traffic signal 30 are calculated. That is, the demand rate ρi of each present i is calculated based on the arrival traffic flow predicted by the arrival traffic flow prediction unit 110, the outflow traffic flow predicted by the outflow traffic flow prediction unit 140 described later, and the like. The demand rate λ at the intersection is calculated by adding the demand rates ρi. Next, the cycle length C is calculated according to the equation (1) based on the calculated intersection demand rate λ. Then, based on the calculated cycle length C and the demand rate ρi of each indication i, a time (CL) obtained by subtracting the loss time L of the own intersection from the cycle length C is used as the demand rate ρi of each indication i. Allocate proportionally and calculate the split.

  The control parameter calculated by the control parameter changing unit 130 is stored as signal control parameter data 336. The control parameter changing unit 130 repeatedly executes this signal control parameter update at predetermined time intervals. Therefore, the signal control parameter is updated at predetermined time intervals.

  Based on the predicted arrival traffic flow calculated by the arrival traffic flow prediction unit 110 and the signal control parameter calculated by the control parameter change unit 130, the outflow traffic flow prediction unit 140 predicts that the outflow traffic flow will flow out from each inflow path at its own intersection. Calculated traffic flow (predicted outflow traffic flow). Specifically, the number of outflows in each traveling direction from each inflow path at its own intersection is calculated for each time t within the prediction target time range. That is, with reference to the signal control parameter data 336, the present at the prediction target time t is determined, and it is determined whether or not the right to pass is given to each inflow path according to the determined present.

  For the inflow path to which the right of passage is not given, as described with reference to FIG. 6, the number of staying in each lane of the inflow path and the number of outflows in each traveling direction are calculated. That is, each lane of the inflow path is determined according to the rate of progress in each traveling direction based on the number of vehicles that arrive at the inflow path at the predicted time t (arrival number) obtained by referring to the predicted arrival traffic flow data 332. The number of vehicles arriving at (the number of arrivals by lane) is calculated.

  The progress rate in each direction is stored in the progress rate table 323. FIG. 22 shows an example of the data configuration of the progress rate table 323. According to the figure, the progress rate table 323 stores a travel direction 323a and a progress rate 323b in association with each other. The traveling direction 323a has three directions of "straight forward", "left turn", and "right turn" because the self-intersection is a cross intersection. The progress rate 323b is determined so that the sum of all the traveling directions is “1.0”.

Next, the outflow traffic flow prediction unit 140 adds the calculated arrival number by lane to the staying number at the time t− ₁ immediately before the prediction time t for each lane of the inflow path, and the staying of the lane at the prediction time t Calculate the number.

  Here, the calculated staying number is stored in the staying number data 334 and the outflow number is stored in the outflow number data 335 for each inflow path.

  The number of outflows by inflow channel data 335 is data of the number of outflows from each inflow channel at the intersection in each traveling direction. FIG. 23 shows an example of the data configuration of the outflow number data 335 for each inflow channel. According to the figure, the outflow number data 335 for each inflow path is generated for each inflow path at its own intersection, and the outflow number 335b from the inflow path in each traveling direction is calculated for each time 335a within the predicted time range. Stored in association. Although the data structure about the inflow path A is shown in the same figure, it is the same structure also about other inflow paths B-D.

On the other hand, as described with reference to FIG. 7, for the inflow path to which the right of passage is given, the number of staying in each lane of the inflow path and the number of outflows from the inflow path in each traveling direction are calculated. That is, in the same manner as the inflow path to which the right of passage is given, each number of the inflow path is determined according to the progress rate in each traveling direction based on the number of arrivals at the predicted time t obtained by referring to the predicted arrival traffic flow data 332. Calculate the number of arrivals to the lane (number of arrivals by lane). Next, for each lane in the inflow path, the number of arrivals by lane calculated at the time t- ₁ immediately before the prediction time t is added to calculate the number of temporarily staying lanes at the prediction time t. Subsequently, with respect to each lane of the inflow path, with reference to the saturated traffic flow table 324, it is determined whether or not the calculated temporary staying number of the lane exceeds the saturated traffic flow.

  The saturated traffic flow table 324 is a data table that defines the saturated traffic flow at the intersection. FIG. 24 shows an example of the data configuration of the saturated traffic flow table 324. According to the figure, the saturated traffic flow table 324 stores a saturated traffic flow 324c in association with each lane 324b of each inflow channel 324a. The saturated traffic flow 324c is a value per lane.

  For a lane whose temporary staying number does not exceed the saturation traffic flow, the calculated temporary staying number is set as the number of outflows from the lane. Based on the number of outflows, the number of outflows from the lane in each direction of travel is calculated according to the rate of progress in each direction of travel, and the inflow at a time (t + Δt) after a predetermined intersection passage time Δt from the predicted time t. The number of spills from the road in the relevant direction of travel. Further, the staying number of the lane at the predicted time t is set to “0”.

On the other hand, for a lane in which the temporarily staying number exceeds the saturated traffic flow, this saturated traffic flow is set as the number of outflows from the lane. Then, based on the number of outflows, the number of outflows from the lane in each direction of travel is calculated according to the progress rate in each direction of travel, and the calculated number of outflows is calculated as a time ( The number of outflows from the inflow path to the corresponding traveling direction at t + Δt). Further, the number of outflows from the calculated lane is subtracted from the number of staying at the time t- ₁ immediately before the predicted time t of the lane, and the number of staying in the lane at the predicted time t is calculated.

  The signal control unit 150 controls each traffic signal 30 at its own intersection according to the signal control parameter calculated by the control parameter changing unit 130.

  The communication control unit 200 controls communication with other traffic signal control devices 20 and external devices such as the central management device 10. For example, the predicted outflow traffic flow data 331 transmitted from the traffic signal control device 20 at the adjacent intersection is received, or the predicted outflow traffic flow data 331 calculated by the outflow traffic flow prediction unit 140 is used as the traffic signal control at the adjacent intersection. It transmits to each apparatus 20.

  The storage unit 300 stores a system program for the processing unit 100 to control the traffic signal control device 20 in an integrated manner, a program and data for realizing the traffic signal control of the present embodiment, and the processing unit 100. Used as a work area, and temporarily stores calculation results and the like executed by the processing unit 100 according to various programs. The storage unit 300 is realized by various IC memories, a hard disk, a ROM, a RAM, and the like, for example. In the present embodiment, the storage unit 300 includes a traffic signal control program 310, a self-intersection configuration table 321, a traffic jam determination point table 322, a progress rate table 323, a saturated traffic flow table 324, and a traffic jam occurrence presence / absence combination table. 325, a traffic jam parameter table 326, a traffic jam occurrence presence / absence determination table 327, a predicted outflow traffic flow data 331, a predicted arrival traffic flow data 332, a staying number data 334, an outflow-specific outflow number data 335, Signal control parameter data 336, sensing result data 341, traffic jam determination result data 344, traffic jam detection result data 342, traffic jam estimation result data 343, and traffic jam determination result data 344 are stored.

[Process flow]
FIG. 25 is a flowchart for explaining the traffic signal control process. This process is realized by the processing unit 100 executing the traffic signal control program 310. According to the figure, first, the arrival traffic flow predicting unit 110 predicts the arrival traffic flow to the own intersection (step A1). That is, based on the predicted outflow traffic flow data 331 received from the traffic signal control device 20 at the adjacent intersection, a traffic flow (arrival traffic flow) predicted to arrive at each inflow path of the own intersection is calculated, and predicted arrival traffic is calculated. Flow data 332 is generated. A known method can be used for the prediction of the arrival traffic flow. Next, control parameter change processing is executed to change the signal control parameter at the own intersection (step A3).

  FIG. 26 is a flowchart for explaining the control parameter changing process. According to the figure, the congestion occurrence determination unit 120 executes a congestion determination process to determine whether or not a congestion has occurred in each inflow path (step B1).

  FIG. 27 is a flowchart for explaining the congestion occurrence determination process. According to the figure, the congestion occurrence determination unit 120 performs the process of loop A for each inflow path. In the loop A, with reference to the signal control parameter data 336, the switching timing from the present having the right of passage to the target inflow path to the present not present is determined (step C1). Next, with reference to the staying number data 334, the predicted staying number immediately before the determined switching timing is determined (step C3). Then, the predicted congestion length is calculated by multiplying the determined predicted staying number by the predetermined total vehicle length (step 5). Loop A is performed in this way.

  When processing of loop A targeting all inflow paths is performed, the congestion occurrence determination unit 120 subsequently performs processing of loop B targeting each determination point. In loop B, it is determined whether or not the vehicle detector 40 is installed at the target determination point (target point) with reference to the traffic congestion determination point table 322. If the vehicle detector 40 is installed (Step C7: YES), the sensing result data 341 is referred to determine a period during which the inflow path corresponding to the target point has the right to pass, and the time occupation in the determined period An average value of the rates is calculated (step C9). If the calculated average occupancy exceeds a predetermined threshold (step C11: YES), the detection result of occurrence of traffic jam at the target point is set to “present” (step C13), and if the threshold is not exceeded (step C13). C11: NO), the detection result of occurrence of traffic jam at the target point is set to “None” (step C15). On the other hand, if the vehicle detector 40 is not installed at the target point (step C7: NO), the detection result of the presence or absence of traffic jam at the target point is set to “unknown” (step C17).

  Subsequently, the congestion occurrence determination unit 120 determines whether or not the predicted congestion length of the inflow path corresponding to the target point has reached the position of the target point. As a result, if the predicted traffic jam length has reached the target point (step C19: YES), the estimation result of the presence or absence of traffic jam at the target point is “Yes”, and if it has not reached (step C21: NO), the target The estimation result of occurrence / non-occurrence of traffic at the point is set to “None” (step C23).

  Thereafter, according to the traffic jam occurrence determination table 327, the traffic jam presence / absence corresponding to the combination of the traffic jam occurrence detection result and the estimation result at the target point is determined (step C25). Loop B is performed in this way. When the process of Loop B targeting all the determination points is performed, the traffic congestion determination process ends.

  When the traffic jam determination process ends, the control parameter changing unit 130 refers to the traffic jam determination result data 344 to determine whether or not a traffic jam has occurred at each determination point. As a result of the determination, if it is determined that the occurrence of the traffic jam is “present” even at one place (step B3: YES), the combination of the presence / absence of the traffic jam at each judgment point is judged based on the traffic jam judgment result data 344. Then, referring to the traffic jam occurrence combination table 325, a control pattern corresponding to the determined combination is selected (step B5). Then, with reference to the control parameter table 326 at the time of traffic jam, the signal control parameter value corresponding to the selected control pattern is changed (step B7).

  On the other hand, if it is determined that there is no traffic jam at all determination points (step 3B: NO), the control parameter changing unit 130 is based on the predicted arrival traffic flow data 332 and the predicted outflow traffic flow data 331. Then, the demand rate ρi of each indication i is calculated (step B9), and the calculated demand rate ρi is added to calculate the demand rate λ of the intersection (step B11). Next, the cycle length C is calculated based on the calculated intersection demand rate λ (step B13), and the split is calculated based on the cycle length C and the demand rate ρi of each indication i (step B15). Then, the signal control parameter is updated to the calculated cycle length C and split (step B17). When the above processing is performed, the control parameter change processing is completed.

  When the control parameter changing process ends, the outflow traffic flow prediction unit 140 calculates a traffic flow (outflow traffic flow) that is predicted to flow out from each inflow path at the intersection, and generates predicted outflow traffic flow data 331. (Step A5). A publicly known method can be used for the prediction of the outflow traffic flow. Then, the processing unit 100 transmits the generated predicted outflow traffic flow data 331 to each traffic signal control device 20 at the adjacent intersection (step A7). Then, it returns to step A1. The processing unit 100 repeatedly executes the processing of steps A1 to A7 at a predetermined time interval (for example, every 5 seconds).

[Action / Effect]
As described above, according to this embodiment, the autonomous distributed traffic signal control device 20 detects, for each determination point of each inflow path at its own intersection, a detection result signal from the vehicle sensor 40 installed at the determination point. Based on (time occupancy rate), the occurrence of traffic jams is detected, and the occurrence of traffic jams is estimated based on the predicted traffic flow of the corresponding inflow route. Based on these detection results and estimation results, It is determined whether or not a traffic jam occurs at the determination point. Then, the signal control parameter is changed to a traffic jam parameter that is determined in advance in association with a combination of determination results at each determination point. This makes it possible to determine the occurrence of traffic jams and to change the appropriate signal control parameters according to the combination of the occurrence of traffic jams in each inflow route and perform signal control, such as appropriate traffic that can be quickly resolved. Signal control is realized.

[Modification]
The applicable embodiments of the present invention are not limited to the above-described embodiments, and can be appropriately changed without departing from the spirit of the present invention.

(A) Change of signal control parameter For example, in the above-described embodiment, when it is determined that the occurrence of traffic jam is “present”, the signal control parameter is “changed” according to the combination of whether or not the traffic jam occurs. “Correction” may be performed. Specifically, first, a signal control parameter is calculated based on the predicted arrival traffic flow and the like, and then the value of this signal control parameter is corrected and changed according to the determination result of the occurrence of traffic congestion. In other words, the display time of the display for giving the right of passage to the inflow path determined to be “existing” is increased by a predetermined time. The lengthening time may be a fixed time (for example, 3 seconds), or may be determined so as to be proportional to the average occupancy rate or the estimated staying number used to determine whether or not a traffic jam has occurred. In this case, the cycle length C becomes longer as the display time increases. Alternatively, the current split for giving the right of passage to the inflow path determined as “present” where the traffic jam has occurred is increased, and the current split for giving the right of traffic to the inflow path determined to be “no” of the traffic jam is reduced. Change to At this time, the cycle length C may also be changed. In any case, correction is performed so that the display time of the display that gives the right of passage to the inflow path determined to be “present” is relatively long.

Configuration of autonomous decentralized traffic signal control system. An example of an intersection. Installation example of a vehicle detector. An example of the presenting method. Overview of traffic signal control. The principle of outflow prediction for inflow channels without access rights. Principle of outflow prediction for inflow channels with right to pass. Configuration of traffic signal control device. The data structural example of a self-intersection structure table. The data structural example of a traffic congestion determination point table. Data configuration example of predicted outflow traffic flow data. The example of a data structure of prediction arrival traffic flow data. The data structural example of sensing result data. The data structural example of traffic jam detection result data. The data structural example of signal control parameter data. Data configuration example of staying unit data. The data structural example of traffic jam estimation result data. 6 is a data configuration example of a congestion occurrence presence / absence determination table. The data structural example of traffic jam determination result data. The data structural example of the congestion occurrence presence / absence combination table. The data structural example of the control parameter table for traffic jams. The data structural example of a progress rate table. Data configuration example of outflow data for each inflow channel. The data structural example of a saturated traffic flow table. The flowchart of a traffic signal control process. The flowchart of the control parameter change process performed during a traffic signal control process. The flowchart of the traffic congestion determination process performed during a control parameter change process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Autonomous decentralized traffic signal control system 20 Traffic signal control apparatus 100 Processing part 110 Arrival traffic flow calculation part, 120 Congestion generation | occurrence | production determination part 30 Control parameter change part, 140 Outflow traffic flow prediction part, 150 Signal control part 200 Communication control part 300 Storage unit 310 Traffic signal control program 321 Self-intersection configuration table 322 Congestion determination point table 323 Progress rate table 324 Saturated traffic flow table 325 Congestion occurrence presence / absence combination table 326 Congestion control parameter table 327 Congestion occurrence determination table 331 Prediction Outflow traffic flow data, 332 Predicted arrival traffic flow data 334 Staying number data, 335 Outflow number data by inflow channel 336 Signal control parameter data 341 Sensing result data, 342 Congestion detection result data 343 Congestion estimation result data, 3 4-congestion determination result data 30 traffic signals, 40 vehicle detector

Claims (4)

Communication means for transmitting and receiving outflow traffic flow information to and from other traffic signal control devices at other intersections; and arrival traffic flow prediction means for predicting arrival traffic flow using the information of outflow traffic flow at the other intersections; Outflow traffic flow prediction means for predicting outflow traffic flow by predicting the number of outflows from each inflow route to the other road by outflow direction, and controlling the signal at its intersection by varying the signal control parameter based on the prediction result An autonomous decentralized traffic signal control device comprising a signal control means,
Using the arrival traffic flow predicted by the arrival traffic flow prediction means, the staying number prediction means for predicting the number of stays in each future inflow path in a predetermined prediction time range in time series,
Congestion occurrence estimation means for estimating the presence or absence of future traffic congestion at each predetermined determination point determined for each inflow path using the staying number within the predicted time range predicted by the staying number prediction means;
Based on the detection result of the vehicle detector that senses the determination point, a congestion occurrence detecting means for detecting occurrence of the congestion at the determination point;
Further comprising
The signal control means determines the presence or absence of traffic jam for each determination point using the estimation result by the traffic jam occurrence estimation means and the detection result by the traffic jam occurrence detection means, and the determination result of the occurrence of traffic jam by the determination point Changing means for changing the signal control parameter according to the combination of
Traffic signal control device. In the case where there is an indication that gives the right of passage to the inflow route related to the determination point where the occurrence of the traffic jam is determined and a indication that gives the right of passage to the inflow route related to the determination point where the occurrence of the traffic jam is not determined. to, to increase the current-splitting of the former, and, traffic signal control apparatus according to claim 1 for changing the signal control parameters so as to reduce the current-splitting of the latter. The congestion occurrence estimation means determines the current switching timing using the signal control parameter, and uses the staying number immediately before the switching timing among the time-series staying numbers predicted by the staying number prediction means. Estimating the presence or absence of traffic congestion on the inflow channel to which the right of passage is given at the indication immediately before the switching,
The traffic signal control device according to claim 1 or 2 . A communication means for transmitting and receiving outflow traffic flow information to and from other traffic signal control devices at other intersections, predicting arrival traffic flow using the outflow traffic flow information at the other intersection, Traffic in an autonomous decentralized traffic signal control device that predicts outflow traffic flow by predicting the number of outflows by direction of outflow to the road, and controls the signal at its intersection by changing the signal control parameter based on the prediction result A signal control method comprising:
Using the predicted arrival traffic flow, predicting the number of remaining inflow passages in the future in a predetermined prediction time range in time series,
Using the number of stays within the predicted time range predicted, the presence or absence of future traffic congestion at each of the predetermined determination points determined for each inflow path,
Based on the detection result of the vehicle detector that senses the determination point, the occurrence of traffic jam at the determination point is detected,
Using the estimation result and the detection result to determine the presence or absence of traffic congestion for each determination point , and changing the signal control parameter according to the combination of the determination results of traffic congestion for each determination point ,
Traffic signal control method.

Images