JP5741326B2 - Traffic index determination device, traffic index determination method, and traffic index determination program - Google Patents

Description

  The present invention relates to a traffic index determination device, a traffic index determination method, and a traffic index determination program for determining a traffic index used for controlling traffic signals.

  For traffic signal control, traffic indicators that indicate the degree of congestion of roads connected to intersections, such as the number of queues, queue length, traffic jam length, and travel time, are used. For example, Patent Document 1 describes that the congestion length is estimated based on sensor information from a vehicle sensor.

JP-A-8-161686

In order to appropriately generate traffic flow by controlling traffic signals, it is necessary for the traffic index to sufficiently follow fluctuations in traffic flow. However, the calculation of the traffic index involves a certain delay due to data smoothing or periodic tabulation, and therefore there is a limit to tracking.
The present invention has been made in view of such problems, and is a traffic index determination device, a traffic index determination method, and a traffic index that can determine a traffic index by following changes in traffic conditions in a relatively short period of time. Provide a decision program.

  The traffic index determination device provided by the present invention determines a traffic index indicating the degree of congestion of a road connected to an intersection. The traffic index determining device includes means for receiving sensor information from a vehicle detector that is installed on a road and senses a vehicle, means for acquiring vehicle information measured by a vehicle traveling on the road, and a first time Means for smoothing traffic indicators for the first time based on sensor information for and reference vehicle information which is vehicle information acquired after the first time until the next sensor information is obtained Is provided.

  The traffic index determination device smoothes the traffic index for the first time based on the sensor information for the first time, the sensor information for the second time before the first time, and the reference vehicle information. Therefore, the situation after the first time can be more quickly reflected on the trend from the second time to the first time. As a result, it is possible to determine the traffic index following the change in traffic conditions in a relatively short time compared to the case where the traffic index is determined only by the sensor information. Here, the time may be a time point or a time, or may be a time zone. The traffic index is, for example, the queue length or the number of queues.

  The traffic index determination device further includes means for calculating the traffic index from the sensor information, and the means for smoothing the traffic index includes a weight for each of the traffic index for the first time and the traffic index for the second time. Based on the reference vehicle information, the traffic index for the first time can be smoothed based on the traffic index for the first time, the traffic index for the second time, and the weight. When performing time smoothing by load averaging, the delay due to the smoothing can be reduced by adjusting the weight calculated for each of the first time and the second time with reference vehicle information.

  In the traffic index determination device, the means for smoothing the traffic index calculates the change tendency of the traffic index during the smoothing period including the first time and the second time from the reference vehicle information from the traffic index for the first time. The weight for the first time may be made larger than the standard value when the change trend to the value matches, and the weight for the first time may be made smaller than the standard value when the change tendency is different. . When the change trends match, the weight for the first time is made larger than the standard value, thereby speeding up the smoothed value following the traffic index for the first time. On the other hand, when the change tendency is different, the weight for the first time is made smaller than the standard value to delay the smoothed value following the traffic index for the first time. Thereby, the followability of the smoothed value with respect to the actual value can be improved.

  In the traffic index determination device, the means for smoothing the traffic index determines the smoothed value of the traffic index for the first time using the reference vehicle information when the number of acquired reference vehicle information is a certain number or more. And when the number of acquired reference vehicle information is less than a fixed number, you may make it smooth the traffic parameter | index with respect to 1st time, without using reference vehicle information.

  When the penetration rate of in-vehicle devices capable of transmitting vehicle information is low, it is difficult to obtain vehicle information from a sufficient number of vehicles in various regions and time zones. When the number of vehicles from which vehicle information can be obtained is limited, the obtained vehicle information may not always be useful in determining a traffic index. For this reason, a traffic index can be determined more appropriately by using reference vehicle information when the number of obtained reference vehicle information is a fixed number or more.

In another aspect, the present invention provides a traffic index determination method corresponding to the traffic index determination apparatus described above. The traffic index determination method includes a procedure for receiving sensor information from a vehicle sensor installed on a road to sense a vehicle, a procedure for obtaining vehicle information measured by the vehicle, and sensor information for a first time. Based on the sensor information for the second time before the first time and the reference vehicle information which is the vehicle information obtained after the first time until the next sensor information is obtained, And a procedure for determining a traffic index for a certain time.
In another aspect, the present invention provides a traffic index determination program for causing a computer to execute the procedure of the traffic index determination method described above.

  According to the present invention, as described above, it is possible to determine a traffic index by following a change in traffic conditions in a relatively short period of time.

The foregoing and other objects, features, and advantages of the present invention will become more apparent from embodiments described below with reference to the accompanying drawings. In the drawings, the same reference numerals denote the same parts in different drawings.
It is a figure showing the whole traffic signal control system composition concerning an embodiment of the invention. It is a figure which shows the example of a structure around the intersection of a traffic signal control system in detail. It is a figure for demonstrating the structural example of a traffic signal controller. It is a figure for demonstrating the structural example of a vehicle-mounted apparatus. It is a figure for demonstrating the structural example of a central apparatus. It is a road top view for demonstrating the method of calculating | requiring the distance from an intersection to the said vehicle based on the position of the vehicle obtained from vehicle information. It is a figure for demonstrating an example of the data which obtains a queue propagation degree from the speed of a vehicle in order to obtain | require queue length from sensor information. It is a figure for demonstrating the method of calculating queue length from queue propagation degree. It is a figure for demonstrating the example of the setting method of the weight in case the distance from the stop line obtained from vehicle information to a vehicle is larger than the queue length obtained from sensor information. It is a figure for demonstrating the example of the setting method of a weight when the distance from the stop line obtained from vehicle information to a vehicle is smaller than the queue length obtained from sensor information. It is a figure which shows an example of the time change of queue length. It is a graph for demonstrating the example of bit allocation of the various data contained in vehicle information.

  FIG. 1 is a diagram showing an overall configuration of a traffic signal control system according to an embodiment of the present invention. The traffic signal control system 100 includes a traffic signal device 101, a central device 102, a vehicle sensor 104 that senses a vehicle 103, and a wireless communication device 105 that wirelessly communicates with an in-vehicle device mounted on the vehicle 103. The traffic signal control system 100 can collect traffic information from the vehicle detector 104 and the wireless communication device 105 to control the traffic signal device 101 and provide traffic information. The traffic signal device 101 is controlled by setting signal control information. Signal control information is typically cycle length, split, and offset.

  The cycle length is the time required for one cycle of the signal indication, and the split is the ratio of the time occupied by each indication to the cycle length. Here, the term “presentation” refers to the right of traffic assigned to a set of traffic flows at the intersection. The offset is expressed as a relative offset that is a time difference in cycle start timing between the target intersection and the adjacent intersection, or an absolute offset that is a time difference in cycle start timing between the target intersection and the reference intersection. In order to set such signal control information, the traffic signal control system 100 can perform MODERATO (Management by Origin-Destination Adaptation for Traffic Optimization) control and profile control.

The central device 102 can perform MODERATO control and profile control. In MODERATO control, all traffic signals 101 belonging to (in charge of) the network of the central device 102 are macro-controlled. In order to deal with a near-saturated traffic state, signal control information for each traffic signal 101 can be determined using the load factor. In the near-saturated state, the traffic volume increases gradually, the traffic density increases, and it becomes difficult to run and overtake at any speed. For example, in the case of split control, a load factor ratio distribution method is adopted in which the maximum load factor of each inflow path for each indication is obtained for each intersection, and the split normalized by the ratio of the indicated load factor is assigned. . The load factor ρ is defined as the following equation (1).
ρ = (Q (T) + E (T−1)) / s (1)
Here, Q (T) is the inflow flow rate (vehicle / hour) of the vehicle in the collection cycle T, E (T-1) is the number of queues (vehicle / hour) generated in the cycle before the collection cycle T, and s is Saturated traffic flow rate (unit / hour).

  In profile control, some traffic signals 101 belonging to the network of the central device 102 are micro-controlled, and the green signal time for the traffic signals 101 is finely adjusted. In profile control, a change in traffic situation is predicted, and the green signal stop timing is determined so that the delay time due to signal waiting at the target intersection is minimized. A simulation calculation is performed based on stop line arrival profile information and signal control information at a certain intersection, and the fluctuation state of the number E of queues from the present time to the future for one cycle or more is calculated. The central device 102 that performs such control can be installed in a traffic control center or on a road. Each traffic signal 101 can also operate by performing distributed processing. In that case, the traffic signal 101 predicts a change in the traffic situation from the information collected by the traffic signal 101 and delivers the prediction result to the adjacent traffic signal 101 to realize distributed processing.

  In the present embodiment, the traffic index determination device of the present invention is used as the central device 102 in the traffic signal control system 100. The number of queues is used for MODERATO control and profile control. The central device 102 specifies the number of queues using both the sensor information obtained from the vehicle sensor 104 and the vehicle information (probe data) of the vehicle 103 obtained from the wireless communication device 105, and sets the number of queues. Based on this, the target traffic signal 101 is controlled.

  FIG. 2 is a diagram showing an example of the configuration around the intersection of the traffic signal control system in more detail. This example is an example of the vicinity of an intersection where the main roads RM1 and RM2 having a large traffic volume intersect with the slave roads RS1 and RS2 having a small traffic volume. The traffic signal at this intersection includes a traffic light controller 201 installed on each of the main roads RM1, RM2 and secondary roads RS1, RS2, and a traffic signal controller connected to these signal lamps 201 so as to be able to control lighting and extinction. 202. The traffic signal controller 202 receives a signal control command from the central device 102 and, based on the signal control information determined from the signal control command, each signal lamp 201 such as blue, yellow, red and right turn arrows. Turns on / off and blinks.

  The vehicle signal detector 104 and the wireless communication device 105 are communicably connected to the traffic signal controller 202. The traffic signal controller 202 receives sensor information from the vehicle sensor 104. The vehicle sensor 104 detects the vehicle 103 flowing into the intersection and transmits a detection result (sensor information) to the traffic signal controller 202. The vehicle detector 104 can measure traffic information such as the number of passing vehicles, a time occupation rate, and an average passing speed by detecting the vehicle 103 on the road. The vehicle sensor 104 is typically an ultrasonic sensor. However, other types of sensors such as a loop coil sensor and a far infrared sensor may be used. The vehicle sensor 104 is installed, for example, 150 m upstream from the stop line. A plurality of vehicle detectors 104 can be installed upstream from the stop line. In general, a traffic measurement sensor is installed 150 m upstream from the stop line, and a traffic jam measurement sensor is installed at 300 m, 500 m, and 250 m intervals from the stop line.

  Further, the traffic signal controller 202 communicates with the in-vehicle device 203 of the vehicle 103 passing through the intersection using the wireless communication device 105. The wireless communication device 105 receives the vehicle information transmitted from the in-vehicle device 203, traffic information such as traffic jam information, safety information such as the presence of a blind spot vehicle, signal time information, map information, and information on nearby stores. Are transmitted to the in-vehicle device 203. As the wireless communication device 105, a mid-range communication device such as a wireless LAN using the UHF band or the VHF band can be used. However, instead of this, a road communication device having a narrow area communication function such as an optical beacon, a radio beacon, RFID, or DSRC may be used. Traffic information can also be provided by wide area communication such as a mobile phone, PHS, and multiplex FM broadcasting. In two-way wide area communication, vehicle information can be acquired via a mobile phone network, a fixed network, the Internet, etc. even if there is no wireless communication device near the intersection.

  FIG. 3 is a diagram for explaining a configuration example of the traffic signal controller. The traffic signal controller 202 includes a wired communication unit 301, a storage unit 302, a lamp drive unit 303, and a calculation unit 304. As the calculation unit 304, one or a plurality of microcomputers can be used. The arithmetic unit 304 is connected to the wired communication unit 301, the storage unit 302, the lamp driving unit 303, and the wireless communication device 105 via an internal bus and various interfaces, and performs traffic signal control by inputting and outputting to these. The function of the machine 202 is realized.

  The wired communication unit 301 is used to communicate with the central device 102 and the vehicle sensor 104 in the present embodiment. A signal control command and traffic information are received from the central device 102, and sensor information is received from the vehicle sensor 104. In addition, the sensor information received from the vehicle sensor 104 and the vehicle information received from the wireless communication device 105 are transmitted to the central device 102. In communication with the central apparatus 102, a relay apparatus (not shown) may be used. The relay device receives the sensor information from the vehicle sensor 104 indirectly or directly via the traffic signal 101, or receives the vehicle information from the wireless communication device 105. Vehicle information is transmitted to the central device 102.

  For the storage unit 302, various recording media such as a hard disk and a semiconductor memory can be used. The storage unit 302 stores various information received from the wired communication unit 301, vehicle information (including vehicle position data and identification data) received from the wireless communication device 105, and the like.

The lamp drive unit 303 includes a semiconductor relay, and in accordance with a command signal from the calculation unit 304, an AC voltage supplied to each color signal lamp corresponding to each of the blue, yellow, and red lamps of the plurality of signal lamps 201. (AC100V) or DC voltage is turned on / off by the relay.
The calculation unit 304 generates a command signal to the lamp drive unit 303 based on the signal control command from the central device 102 received by the wired communication unit 301, thereby changing the signal lamp color of each signal lamp unit at an appropriate timing. Let them switch.
Information such as vehicle information communicated and recorded by the traffic signal controller may be provided with security for privacy protection in terms of hardware or software.

  FIG. 4 is a diagram for explaining a configuration example of the in-vehicle device. The in-vehicle device 203 communicates with the wireless communication device 105 to exchange various information with the traffic signal controller 202 and to navigate to the destination. The in-vehicle device 203 includes a GPS processing unit 401, a direction sensor 402, a vehicle speed acquisition unit 403, a communication unit 404, a storage unit 405, an operation unit 406, a display unit 407, an audio output unit 408, and a calculation unit 409.

The GPS processing unit 401 receives a GPS signal from a GPS satellite, and measures the position (latitude, longitude, and altitude) of the vehicle based on time information included in the GPS signal, the orbit of the GPS satellite, positioning correction information, and the like. .
An optical fiber gyro can be used for the orientation sensor 402. The direction sensor 402 measures the direction and angular velocity of the vehicle. The vehicle speed acquisition unit 403 acquires vehicle speed data measured by the vehicle speed sensor detecting the angular velocity of the wheels.

  The communication unit 404 can perform communication with a vehicle-mounted device on another vehicle, that is, vehicle-to-vehicle communication, during a period when wireless transmission by the wireless communication device 105 is not performed. The communication unit 404 transmits the identification data of the own device, the position data of the own vehicle, the speed data, and the like as vehicle information in real time (for example, at a cycle of 0.1 to 1.0 seconds). This vehicle information is monitored by the wireless communication device 105. The communication unit 404 can also receive traffic information from the wireless communication device 105 in the communication area of the wireless communication device 105 by traveling of the vehicle.

  Various storage media such as a hard disk and a semiconductor memory can be used for the storage unit 405. The storage unit 405 stores traffic information received by the communication unit 404, vehicle information data measured by the in-vehicle device 203 itself, and the like. The storage unit 405 may store vehicle information of surrounding vehicles obtained by inter-vehicle communication. The stored vehicle information data is output to the communication unit 404 at any time by the calculation unit 409 and transmitted from the communication unit 404. When spot communication is performed like an optical beacon, the calculation unit 409 reads the vehicle information data stored in the storage unit 405 in the spot communication area, and the communication unit 404 uplinks the data. The storage unit 405 also stores a map database.

  The road map data in this map database includes the intersection data in which the intersection ID and the position of the intersection are associated, the link ID, the start point / end point / interpolation point of the link (corresponding to the point where the road bends), the link Link data including the link ID of the link connected to the start point of the link, the link ID of the link connected to the end point of the link, and the link cost. For example, there are as many link costs as the number of combinations of a link and its end point, and after entering the start point of the link, exit the end point of the link and enter the start point of the next link to be connected. It is determined as the time required for That is, the link cost includes the cost (time) required to travel from the start point to the end point of the link and the cost (time) required to travel from the end point of the link to the start point of the next link, that is, the intersection. Includes time taken to pass.

  The operation unit 406 includes a touch panel and buttons so that a vehicle passenger including a driver can set a destination. The display unit 407 is a monitor device attached to a dashboard portion of the vehicle, for example, and displays an interface screen. The audio output unit 408 outputs the audio data created by the calculation unit 409 from the speaker.

  As the calculation unit 409, one or a plurality of microcomputers can be used. The calculation unit 409 is connected to the GPS processing unit 401, the direction sensor 402, the vehicle speed acquisition unit 403, the communication unit 404, the storage unit 405, the operation unit 406, the display unit 407, and the audio output unit 408 via a bus. The functions of the in-vehicle device 203 are realized by performing input / output.

  The calculation unit 409 stores the vehicle position measured by the GPS processing unit 401, the vehicle azimuth and angular velocity measured by the azimuth sensor 402, the vehicle speed data acquired by the vehicle speed acquisition unit 403, and the storage unit 405. Map matching processing is performed based on the road map data to determine the position of the vehicle on the road map data link. When the setting of the destination is received by the operation unit 406, the calculation unit 409 calculates a route to the destination for the vehicle position, and uses the display unit 707 and the audio output unit 408 to calculate the route. Guide the results.

  FIG. 5 is a diagram for explaining a configuration example of the central apparatus. One or more computers can be used for the central device 102. In the example of FIG. 5, one computer is used as the central device 102. The central apparatus 102 includes a storage unit 501, a communication unit 502, a display unit 503, an operation unit 504, and a calculation unit 505. The calculation unit 505 is connected to the storage unit 501, the communication unit 502, the display unit 503, and the operation unit 504 via a bus, and realizes the function of the central device 102 by performing input / output to these units. .

  As the storage unit 501, a computer-readable medium such as a nonvolatile ROM, a flash memory, various semiconductor memories such as a volatile RAM, and a hard disk can be used. The storage unit 501 stores control programs for performing MODERATO control and profile control, execution files of other programs, data of various traffic indexes used for this control, and the like. The execution file includes an execution file or program module of a traffic index determination program describing each procedure of the traffic index determination method of the present invention.

  When the calculation unit 505 executes the traffic index determination program, the computer used for the central apparatus 102 can perform each procedure of the traffic index determination method. Thereby, the central apparatus 102 functions as a traffic index determination apparatus of the present invention. Instead of or in addition to using the general-purpose computer and the traffic index determination program, part or all of the procedure of the traffic index determination method or the function of the traffic index determination device may be realized by dedicated hardware.

  The traffic index determination program can be installed in the storage unit 501 using a telecommunication line or a computer-readable medium. For example, a removable semiconductor memory device having an optical disk such as a CD, DVD, or BD, a magnetic disk such as a flexible disk, or a flash memory can be used for the introduction.

  The communication unit 502 transmits traffic information including traffic information and signal control commands to each traffic signal every predetermined time. The signal control command is transmitted every signal control information calculation cycle (for example, 1.0 to 2.5 minutes), and the traffic information is transmitted, for example, every 5 minutes.

  The communication unit 502 receives, from each traffic signal, vehicle information that is information related to the vehicle on which the in-vehicle device is mounted, and sensor information of the vehicle sensor 104 that includes a pulse signal generated when the vehicle passes through in real time. . The vehicle information includes at least vehicle position data and identification data. It may further include vehicle speed data, vehicle direction, operation information of the driving vehicle, and the like.

  When the central device 102 functions as the traffic index determination device of the present invention, the communication unit 502 receives vehicle information including means for receiving sensor information from the vehicle sensor 104 and vehicle position data measured by the vehicle. Acts as a means to obtain. Here, “measured by the vehicle” is not limited to the case of measurement by the vehicle itself, but includes the case of measurement by a device installed or brought into the vehicle, such as an in-vehicle device or a mobile phone.

  The display unit 503 displays a road map of the area managed by the central device 102 and a screen including the positions of all traffic signals and other road equipment on the road map, and notifies the operator of the situation such as traffic jams and accidents. Used to do. The operation unit 504 is an input interface such as a keyboard and a mouse. The operation unit 504 allows the operator to perform a display switching operation on the display unit 503.

  When the central device 102 functions as the traffic index determination device of the present invention, the calculation unit 505 detects sensor information for the first time, sensor information for the second time before the first time, and the first time. It operates as a means for determining a traffic index for the first traffic time based on reference vehicle information, which is vehicle information acquired after the time until the next sensor information is obtained. Here, the traffic index is the queue length or the number E of queues used for MODERATO control or profile control.

FIG. 6 is a road plan view for explaining a method for obtaining the distance from the intersection to the vehicle based on the position of the vehicle obtained from the vehicle information. This figure shows a state in which a vehicle 601 equipped with an in-vehicle device is stopped upstream of an intersection due to waiting for a signal. Vehicle information can be recorded periodically or at a stop. If the vehicle information received from the in-vehicle device of the vehicle 601 includes a stop position in the case of waiting for a signal, the central device 102 uses the stop position included in the vehicle information, and from there, an intersection in the map database it can determine the distance L ₁ to C node. The central device 102 can determine the distance L _p from the stop line to the vehicle by reducing the distance L ₂ (constant) from the stop line to the node with respect to the distance L ₁ . If stopped at the end of the vehicle 601 the queue, the queue size wait that distance L _p. Dividing that value by the average vehicle head gap H at the time of stopping gives the number of queues.

  However, the target vehicle 610 is not always stopped at (at the end of) the queue. When there are few vehicles that transmit vehicle information, if the queue length and the number of queues are determined using this vehicle information, there is a possibility that the vehicle information will deviate greatly. That is, when there are few vehicles that transmit vehicle information, the reliability of the queue length obtained from the vehicle information becomes low.

  However, since the vehicle information represents immediate information, the vehicle information represents a tendency that is more recent than the sensor information that performs periodic counting. If vehicle information indicating that the vehicle has stopped at a position farther from the stop line than the queue length obtained from the sensor information can be obtained, the reliability of the queue length calculated from the vehicle information is low. It can be determined that the queue length measured by the vehicle sensor is longer. That is, for example, when smoothing the queue length, the weighting factor is increased for the new queue length, and the weighting factor is decreased for the old queue length. Can be performed.

FIG. 7 is a diagram for explaining an example of data for obtaining the queue spread from the vehicle speed in order to obtain the queue length from the sensor information. This data is data representing the relationship between the speed of the vehicle and the spillover degree of the queue, and includes, for example, speed thresholds s1, s2, and s3 set in advance for each vehicle sensor. The vehicle speed S [m / sec] is given by the following equation (2) using the inflow flow rate Q [vehicles / sec], the occupation rate O, and the effective vehicle length (set value) Le [m].
S = (Le · Q) / O (2)
Here, the effective vehicle length Le is a value obtained by adding the sensing region [m] to the average vehicle length [m].

  The arithmetic unit of the central device determines the flow rate Q and the occupation rate O from the sensor information, reads the average vehicle length and the sensing area from the storage unit, calculates the effective vehicle length Le, the flow rate Q, the occupation rate O, Then, the vehicle speed is calculated from the effective vehicle length Le according to the above equation (2). From the speed of the vehicle, a queue propagation degree indicating the degree of propagation of the queue to the vehicle detector is determined. A value of 0 to 1.0 is given to the queue spread depending on the speed of the vehicle.

The speed threshold s1 corresponds to a queue ripple degree of 1.0 and represents the speed in a state where the end of the queue is sufficiently beyond the vehicle detector position. The speed threshold s2 corresponds to the queue ripple degree of 0.5, and represents the speed in a state where the end of the queue exists near the vehicle detector position. The speed threshold s3 corresponds to a queue ripple degree of 0 and represents the speed at which the queue does not affect the sensor speed. The data stored in the storage unit of the central device includes, for example, the speed threshold values s1, s2, and s3, and the value of the queue spread for these speed threshold values. Using these speed thresholds s1, s2, and s3, the queuing ripple r is given by the following equation (3) based on the average speed S obtained by the vehicle sensor.
r = 1 (S <s1)
(S2-S) / (s2-s1) (s1 <S <s2) (3)
(S3-S) / (s3-s2) * 0.5 (s2 <S <s3)

  FIG. 8 is a diagram for explaining a method of calculating the queue length from the queue spread degree. FIG. 8 shows an example of the relationship between sensor position and queue spread. This example includes queue propagation for each of the three vehicle detectors D1, D2, and D3. When each point is connected with a straight line, the point where the queue ripple is less than 0.5 indicates the end of the traffic jam, and in the section where the queue ripple is continuously 0.5 or more downstream from the traffic jam tail The length indicates the queue length. The calculation unit of the central device calculates a queue length by calculating a straight line connecting the points from the queue spread for each vehicle sensor and calculating a position where the queue spread is 0.5.

In the present embodiment, the arithmetic unit of the central device determines the queue length by smoothing the queue length thus obtained. This is to eliminate variations in time series data. Various methods such as moving average and exponential smoothing can be used for smoothing. Smoothing may be performed in the frequency domain. Here, the queue length Ls (T) after smoothing is given by the following equation (4).
Ls (T) = αL (T) + (1−α) L (T−1) (4)
Here, the weight α is larger than 0 and smaller than 1.

  The arithmetic unit of the central device acquires the vehicle between the collection period T of the sensor information and (T-1) one cycle before, or between the collection period T and (T + 1) after one cycle. The weight α can be adjusted based on the reference vehicle information that is information. Even if vehicle information is available, it is not necessary to use the vehicle information at any time. For example, if the stop position (distance from the stop line) of the vehicle obtained from the vehicle information is smaller than a threshold value, it is estimated that there is no traffic jam or the traffic jam length is short. In such a case, it is not necessary to adjust using the vehicle information.

  Here, the calculation unit is a stop line obtained from the relationship between the queue lengths L (T-1) and L (T) and the threshold Lc, and the queues L (T-1) and L (T) and the vehicle information. The weight α is adjusted based on the relationship with the distance Lp (Tp) from the vehicle to the vehicle. The threshold value Lc is, for example, 150 [m], and corresponds to the distance from the stop line to the vehicle detector.

  When the queue lengths L (T-1) and L (T) are equal to or smaller than the threshold value Lc, the arithmetic unit of the central device sets the weight α according to the stop time of the vehicle with respect to the red signal. For example, when the vehicle stops after a considerable time has elapsed since the start of the red signal, the congestion length should be short, and the vehicle information and the queue length tend to be the same, so there is no need to adjust the weight α, α may be set to a value (for example, 0.8) larger than a standard value (for example, 0.5). The red signal start time can be obtained from the signal control information.

When the queue lengths L (T-1) and L (T) are larger than the threshold value Lc, the arithmetic unit of the central device determines that the distance Lp (Tp) obtained from the vehicle information is expressed by the following equation (5) or (6): Determine whether the relationship is satisfied.
L (T-1) <Lp (Tp) <L (T) (5)
L (T) <Lp (Tp) <L (T-1) (6)
When the relationship of the above formula (5) or (6) is satisfied, both the change in the queue length and the value obtained from the vehicle information indicate that the queue length is monotonically increasing or monotonically decreasing during one cycle. Show. That is, the vehicle information indicates a value that matches the change in the queue length obtained from the vehicle detector. For this reason, the calculation unit sets the weight α to a standard value (for example, 0.5) and does not adjust the weight α.

Furthermore, the arithmetic unit adjusts the weight α when the expression (5) or (6) is not satisfied and the following expression (7) or (8) is satisfied. At this time, the vehicle information indicates that the change in the queue length obtained from the vehicle detector is delayed from the actual change.
Max (L (T-1), L (T)) <Lp (7)
Min (L (T-1), L (T))> Lp (8)
When Expression (7) is satisfied and L (T)> L (T−1), the calculation unit sets the weight α to a value larger than the standard value (for example, 0.7). When the above equation (7) is satisfied and L (T-1)> L (T), the calculation unit sets the weight α to a value smaller than the standard value (for example, 0.3). When the above equation (8) is satisfied and L (T)> L (T-1), the calculation unit sets the weight α to a value smaller than the standard value, and L (T−1)> L (T) In this case, the weight α is set to a value larger than the standard value.

  In this way, the arithmetic unit of the central device reflects the vehicle information acquired between the sensor information collection period T and (T-1) one cycle before that in the smoothing weight α, thereby smoothing. The followability of the queue length Ls after conversion can be improved. The calculation unit may correct the queue length Ls (T) by adjusting the weight α based on the vehicle information acquired between the collection period T and (T + 1) after one cycle. By correcting the queue length Ls (T) before calculating the queue length Ls (T + 1), the followability can be improved as a result.

  For example, when the distance Lp (Tp) from the stop line to the vehicle obtained from the vehicle information is larger than the queue length L (T) obtained from the sensor information, the arithmetic unit may calculate the queue length L (T) and Of the queue length L (T−1), the weight α for the larger value is made larger than the standard value (for example, 0.5).

  FIG. 9 is a diagram for explaining an example of a weight setting method when the distance from the stop line obtained from the vehicle information to the vehicle is larger than the queue length obtained from the sensor information. FIG. 9A shows an example where the value of the queue length L (T) for cycle T is greater than the queue length L (T−1) for cycle (T−1), and FIG. 9B shows the queue length L (T). Shows an example in which the value of is less than or equal to the queue length L (T-1). In the case of the example of FIG. 9A, the arithmetic unit of the central device increases the weight for the queue length L (T) and decreases the weight for the queue length L (T−1). That is, the weight α in the equation (4) is set to a value larger than the standard value, for example, 0.7. As a result, the value of the smoothed queue length Ls (T) becomes close to the queue length L (T). Therefore, the followability is improved as compared with the case where the weight α is kept at the standard value. Further, the value of the distance Lp (Tp) is only used for adjusting the weight α, and the queue length Ls (T) after smoothing obtained from the above equation (4) is the queue length before smoothing. L (T) is not exceeded. Thereby, overtracking is suppressed. As a result, in this example, the followability can be improved while maintaining smoothness of about one cycle. In the case of the example of FIG. 9B, the arithmetic unit of the central device decreases the weight for the queue length L (T) and increases the weight for the queue length L (T−1). That is, the weight α is set to a value smaller than the standard value, for example, 0.3. This delays the smoothed value Ls (T) from following the newest queue length L (T) obtained from the sensor information. As a result, even when the actual queue length changes from a decreasing tendency to an increasing tendency around the cycle T, the value after smoothing quickly follows the increasing tendency.

  FIG. 10 is a diagram for explaining an example of a weight setting method when the distance from the stop line obtained from the vehicle information to the vehicle is smaller than the queue length obtained from the sensor information. FIG. 10A shows an example where the value of the queue length L (T) is larger than the queue length L (T−1), and FIG. 10B shows the value of the queue length L (T) being the queue length L (T− 1) An example in the following case is shown. In the example of FIGS. 10A and 10B, the arithmetic unit of the central device sets the weight α to a standard value. In the example of FIG. 10A, although the distance Lp (Tp) is smaller than the queue length L (T), the follow-up to the queue length L (T-1) is not delayed. This is because the received vehicle does not always stop at the end of the queue, and the true value L (Tp) of the queue length at time Tp may be larger than the distance Lp (Tp). In the example of FIG. 10B, the distance Lp (Tp) is smaller than the queue length L (T), but the follow-up to the queue length L (T-1) is not accelerated. This is because the true value L (Tp) of the matrix length may be larger than the distance Lp (Tp).

  However, also in the examples of FIGS. 10A and 10B, the arithmetic unit of the central device may adjust (change) the weight α from the standard value. This is because the distance from the stop line to the vehicle obtained from the vehicle information does not greatly deviate from the actual queue length when the in-vehicle device becomes popular. In the example of FIG. 10A, the arithmetic unit decreases the weight for the queue length L (T) and decreases the weight for the queue length L (T−1). That is, the weight α is set to a value smaller than the standard value, for example, 0.3. This delays the smoothed value Ls (T) from following the newest queue length L (T) obtained from the sensor information. As a result, even if the actual queue length changes from an increasing tendency to a decreasing tendency around the cycle T, the value after smoothing quickly follows the decreasing tendency. In the case of FIG. 10B, the calculation unit increases the weight for the queue length L (T) and decreases the weight for the queue length L (T−1). That is, the weight α is set to a value larger than the standard value, for example, 0.7. This speeds up the smoothed value Ls (T) following the newest queue length L (T) obtained from the sensor information. Further, the value of the distance Lp (Tp) is only used for adjusting the weight α, and the queue length Ls (T) after smoothing obtained from the above equation (4) is the queue length before smoothing. Not below L (T). Thereby, overtracking is suppressed. As a result, in this example, the followability can be improved while maintaining smoothness of about one cycle.

  As described above, the calculation unit of the central device can improve the followability by reflecting the latest information obtained from the vehicle information in the smoothing weight α. In the example of FIGS. 9 and 10, the weight α is adjusted in two stages, but it may be performed in three stages or more or continuously. For example, the weight α is changed according to the difference between the distance Lp (Tp) from the stop line to the vehicle obtained from the vehicle information and the queue length L (T) obtained from the sensor information. be able to. Further, in the example of FIG. 9 (and FIG. 10), the weight α is adjusted from the standard value when there is a difference between the distance Lp (Tp) and the queue length L (T), but the distance Lp (TP) And the queue length L (T) may be adjusted when the difference α exceeds a predetermined value. In this case, if the difference between the distance Lp (Tp) and the queue length L (T) is smaller than a predetermined value, the weight α is set to a standard value.

  Furthermore, in the example of the above formula (4), the queue length L (T) is smoothed using the past one sample L (T-1), but smoothing is performed using a plurality of past samples. Also good. In this case, the arithmetic unit of the central device, when the change tendency within the smoothing period or the smoothing period and the change tendency from the queue length L (T) to the distance Lp (Tp) coincide, The weight for L (T) is made relatively large (the weight for other samples in the smoothing period is made small), the change tendency in the smoothing period or the smoothing period, and the queue length L When the change tendency from (T) to the distance Lp (Tp) is different, the weight for the queue length L (T) is relatively reduced. Judgment whether or not the change trend is the same as long as the slope is positive or negative, for example, it is not required for that judgment until the slope value matches. As the value of the past sample within the smoothing period or the smoothing period, a value after smoothing may be used, or a value before smoothing may be used.

  Further, the arithmetic unit of the central device adjusts the weight α based on the reference vehicle information when the number of vehicles from which the reference vehicle information is obtained is a certain value or more, and the number of vehicles from which the reference vehicle information is obtained is set to a constant value. If not, the queue length may be smoothed without adjusting the weight α. Furthermore, only when the number of vehicles for which reference vehicle information can be obtained is greater than or equal to a certain value, the distance from the stop line obtained from the vehicle information to the vehicle is greater than the queue length obtained from the sensor information. In this case, the weight α may be adjusted.

  FIG. 11 is a diagram illustrating an example of a temporal change in the queue length. In this figure, the solid line represents the queue length obtained only from the sensor information, the dotted line represents the actual queue length, and the broken line represents the queue length determined as in the present embodiment. The chain line represents the queue length obtained from the vehicle information. Due to the periodic counting of the sensor information, the queue length obtained only from the sensor information has a time delay Δ with respect to the actual queue length. Although the queue length obtained from the vehicle information closely follows the actual queue length, sufficient accuracy cannot be secured for the value. When the weight is adjusted as in the present embodiment, the actual value is followed in a relatively short time while ensuring the necessary accuracy as the queue length value.

  In the present embodiment, vehicle information is preferably transmitted in real time in order to increase the responsiveness of the queue length. However, a near real time probe such as an optical beacon may be used. This is because there is a collection delay until the vehicle passes the light beacon point, but information may be obtained earlier than the vehicle detector.

  FIG. 12 is a chart for explaining an example of bit allocation of various data included in the vehicle information. When the vehicle information is accumulated and transmitted from the in-vehicle device in the spot communication area as in the case of the optical beacon, the vehicle information having the data configuration as in this example can be used. In this example, single stop, direction change, constant distance travel / direction change, and uplink are prepared as types of events in the vehicle. Data to be transmitted from the in-vehicle device can be selected according to the event type. In case of single stop, stop time data, direction change, absolute direction data, fixed distance travel / direction change, travel distance from previous event, uplink, invalid value are recorded respectively. The The stop time in the case of single stop is represented by 8 bits, and the case of second and minute is distinguished by the value of the first bit, and the remaining 7 bits represent time. For this reason, the time from 0x01 (1 second) to 0xff (127 minutes) can be allocated in hexadecimal with 1 second as the minimum unit. Further, the absolute direction in the case of the direction change is assigned as 16 units in the clockwise direction with “1” in the north. In the case of constant distance running or direction change, 8 bits are assigned to the distance from the previous event, and 1 bit is in units of 5 m. In this case, a travel distance from 0x01 (5 m) to 0xff (1275 m) can be assigned in hexadecimal. For example, by using the contents of these events (stop time, absolute displacement, travel distance), the usefulness can be similarly determined for the accumulation type vehicle information. Although not shown in FIG. 10, the number of stops until the vehicle passes through the intersection can be recorded as one information item and used for determining the usefulness of the vehicle information.

  In the embodiment described above, the traffic index determination device of the present invention is used as a central device in a traffic signal control system, but the present invention is not limited to this. For example, in the examples of FIGS. 9 and 10, the queue length is obtained from the sensor information and the queue length is smoothed. However, the vehicle speed obtained from the sensor information is smoothed, and the queue is obtained from the smoothed vehicle speed. The queue length may be smoothed by obtaining the length. That is, by smoothing the traffic index at different times, the traffic index is not smoothed, but the quantity that is the basis of the traffic index is smoothed, and the traffic index is obtained from the quantity, thereby obtaining the traffic index. The index can be smoothed. Moreover, although the value of 0.3, 0.5, and 0.7 was illustrated as an example of the weight at the time of smoothing, it is not restricted to this. The weight may be an extremely small or large value, 0.1 or 0.9, and may be 0 or 1.

  Moreover, you may make it utilize the traffic parameter | index determination apparatus of this invention for a traffic signal controller and other apparatuses. Furthermore, the traffic index determination device may be configured by a plurality of devices in hardware. Furthermore, the present invention can be used not only when a traffic index is used for traffic signal control but also when a traffic index is calculated for driving a vehicle. A typical example of the traffic index to be calculated is the queue length or the number of queues. However, the present invention can also be used when calculating queue lengths, load factors, travel times, and other traffic indicators that indicate the degree of congestion on roads connected to intersections.

  The above-described embodiments do not limit the technical scope of the present invention, and various modifications and applications are possible within the scope of the present invention.

  The traffic index determination device, the traffic index determination method, and the traffic index determination program according to the present invention can determine a traffic index following a change in traffic conditions in a relatively short period of time, such as traffic signal control and provision of traffic information. It can be used for various purposes.

DESCRIPTION OF SYMBOLS 100 Traffic signal control system 101 Traffic signal machine 102 Central apparatus 103 Vehicle 104 Vehicle detector 105 Wireless communication apparatus 201 Signal lamp 202 Traffic signal controller 203 Car-mounted apparatus 301 Wired communication part 302 Storage part of traffic signal controller 303 Lamp drive part 304 Operation unit of traffic signal controller 401 GPS processing unit 402 Direction sensor 403 Vehicle speed acquisition unit 404 In-vehicle device communication unit 405 In-vehicle device storage unit 406 In-vehicle device operation unit 407 In-vehicle device display unit 408 Audio output unit 409 In-vehicle device Arithmetic unit 501 Central unit storage unit 502 Central unit communication unit 503 Central unit display unit 504 Central unit operation unit 505 Central unit arithmetic unit 601 Vehicle

Claims (7)

A traffic index determination device that determines a traffic index indicating a degree of congestion of a road connected to an intersection,
Means for receiving sensor information from a vehicle sensor installed on the road for sensing the vehicle;
Means for obtaining vehicle information measured by a vehicle traveling on the road;
The sensor information for a first time, the sensor information for a second time prior to the first time, and acquired between the first time and the second time, or Means for smoothing the traffic index for the first time based on reference vehicle information, which is the vehicle information acquired until the next time the sensor information is obtained after the first time. Traffic index determination device. Means for calculating the traffic index from the sensor information;
The means for smoothing the traffic indicator is:
Calculating a weight for each of the traffic index for the first time and the traffic index for the second time based on the reference vehicle information;
The traffic index determination device according to claim 1, wherein the traffic index for the first time is smoothed based on the traffic index for the first time, the traffic index for the second time, and the weight. The means for smoothing the traffic indicator is:
The change tendency of the traffic index in the smoothing period including the first time and the second time coincides with the change tendency from the traffic index to the value calculated from the reference vehicle information with respect to the first time. 3. The traffic index determination device according to claim 2, wherein the weight for the first time is made larger than a standard value, and when the change tendency is different, the weight for the first time is made smaller than the standard value. . The means for smoothing the traffic indicator is:
When the number of the acquired reference vehicle information is a predetermined number or more, the smoothed value of the traffic index for the first time is determined using the reference vehicle information, and the number of the acquired reference vehicle information The traffic index determination device according to any one of claims 1 to 3, wherein the traffic index for the first time is smoothed without using the reference vehicle information when the predetermined number is not satisfied.   The traffic index determination device according to claim 1, wherein the traffic index is a queue length or the number of queues. A traffic index determination method for determining a traffic index indicating a degree of congestion of a road connected to an intersection,
Receiving detector information from a vehicle detector installed on the road for detecting the vehicle;
Obtaining vehicle information measured by a vehicle traveling on the road;
The sensor information for a first time, the sensor information for a second time before the first time, and the next time the sensor information is obtained after the first time. And a procedure for smoothing the traffic index for the first time based on reference vehicle information which is the vehicle information.   A traffic index determination program for causing a computer to execute the procedure according to claim 6.

Images