Traffic information generating device, traffic information generating method, and traffic information generating program

Abstract

To improve the accuracy of traffic information on an observation target area generated based on a result of detection of vehicles performed by a radar device.SOLUTION: A detection data acquisition unit acquires detection data of vehicles traveling in an observation target area, the data detected by a radar device scanning the observation target area by a survey wave. A generation unit processes the detection data of vehicles acquired by the detection data acquisition unit, and for each of a plurality of sections obtained by dividing the observation target area in the direction of travel of the vehicles, generates sectional traffic information indicating the state of traffic flow of vehicles in the section. An integration unit integrates the pieces of sectional traffic information generated by the generation unit for the sections to generate traffic information on the observation target area.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a technique for detecting vehicles traveling in an observation area and generating traffic information for the observation area. [Background technology]

[0002] Conventionally, an observation area is repeatedly scanned with a probe wave to detect vehicles traveling in the observation area (see, for example, Patent Document 1). The observation area is, for example, a road (highway, general road) on which vehicles travel. The probe wave is, for example, a millimeter wave (radio wave) or laser light. A device that uses millimeter waves as the probe wave is generally called a radio wave radar, and a device that uses laser light as the probe wave is generally called a laser radar. Here, radio wave radar and laser radar are collectively referred to as radar devices.

[0003] There is also a traffic information generating device that generates traffic information showing the traffic flow situation in an observation area based on the detection results of vehicles traveling in the observation area.The traffic information generating device generates traffic information such as the travel time in the observation area (the time from when a vehicle enters the observation area to when it leaves) and the average speed of vehicles in the observation area. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6716956 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the further away a radar device is from the position where the search wave is emitted (the installation position of the radar device), the lower the vehicle detection accuracy (the more frequently vehicles are missed from detection). For example, if a radar device is installed downstream of an observation area, the vehicle detection accuracy on the upstream side of the observation area will be lower. Conversely, if a radar device is installed upstream of an observation area, the vehicle detection accuracy on the downstream side of the observation area will be lower.

[0006] In radar devices, factors that cause a decrease in vehicle detection accuracy are thought to be, for example, the fact that the irradiation density of the probe wave becomes coarser the farther away from the probe wave irradiation position, and the fact that the further a vehicle is located from the probe wave irradiation position, the more likely it is that the probe wave will not be irradiated because it is hidden by a vehicle located closer to the probe wave irradiation position (the vehicle in front or the vehicle behind).

[0007] The traffic information generating device generates traffic information based on detection data of vehicles continuously detected by the radar device from their entry into the observation area to their exit. In other words, the traffic information generating device did not use detection data of vehicles that were not detected entering the observation area or leaving the observation area due to detection failures in generating traffic information.

[0008] Furthermore, if a detection miss occurs in the middle of an observation area, it is difficult to match the detection result before the detection miss and the detection result after the detection miss as the detection result of the same vehicle. In other words, if a detection miss occurs in the middle of an observation area, the detection result before the detection miss and the detection result after the detection miss are likely to be matched to different vehicles, reducing the accuracy of the vehicle detection result. The reduction in the accuracy of the vehicle detection result reduces the accuracy of the traffic information generated based on this detection result. For this reason, in the past, if a detection miss occurs in the middle of an observation area, the detection result before the detection miss and the detection result after the detection miss were considered to be the detection results of different vehicles and were not used to generate traffic information.

[0009] This reduces the number of vehicle detection results that the traffic information generating device uses to generate traffic information for the observation area, lowering the accuracy of the traffic information generated based on the vehicle detection results.

[0010] An object of the present invention is to provide a technique for improving the accuracy of traffic information for an observation target area that is generated based on the results of vehicle detection by a radar device. [Means for solving the problem]

[0011] In order to achieve the above object, the traffic information generating device of the present invention is configured as follows.

[0012] The detection data acquisition unit acquires detection data of vehicles traveling in the observation area, which are detected by the radar device scanning the observation area with a search wave. The radar device may be, for example, a radio wave radar that irradiates millimeter waves (radio waves) as the search wave, or a laser radar that irradiates laser light as the search wave.

[0013] The generation unit processes the vehicle detection data acquired by the detection data acquisition unit, and generates section traffic information indicating the state of vehicle traffic flow in each of a plurality of sections obtained by dividing the observation target area in the vehicle travel direction. For example, the generation unit generates section traffic information such as section travel time and section average speed for each section obtained by dividing the observation target area in the vehicle travel direction.

[0014] The integration unit integrates the section traffic information generated for each section by the generation unit, and generates traffic information for the observation target area.

[0015] This configuration prevents a decrease in the number of vehicles 110 from which detection data used to generate section traffic information for each section can be obtained, thereby improving the accuracy of traffic information for the observation target area.

[0016] Furthermore, for example, the integrating unit may generate the sum of the section travel times generated by the generating unit for each section as the travel time for the observation target area. Furthermore, if the observation target area is an entrance road to an intersection, the generating unit may integrate the section travel times generated for each section, taking into account stops due to red lights of traffic lights installed at the intersection, stops due to congestion, etc.

[0017] Furthermore, for example, the generation unit extracts vehicles that can be tracked from the start point to the end point of each section obtained by dividing the observation area in the vehicle travel direction, and generates the section travel time for that section using the tracking time during which each extracted vehicle was detected in that section. The section travel time may be the average value of the time required for vehicles to travel that section, or it may be the median, mode, root mean square, etc.

[0018] In addition, the generation unit may generate the travel time for each section obtained by dividing the observation area in the vehicle's traveling direction, using the vehicle's stopped state and tracking time in that section detected based on the vehicle detection data acquired by the detection data acquisition unit. [Effects of the Invention]

[0019] According to the present invention, it is possible to improve the accuracy of traffic information for an observation target area that is generated based on the results of vehicle detection by a radar device. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a schematic diagram showing an example of an observation target area for which the traffic information generating device of this example generates traffic information. [Figure 2] 1 is a schematic diagram showing an example of an area that the traffic information generating device of this example covers with a probe wave to generate traffic information. FIG. [Figure 3] FIG. 10 is a diagram illustrating an example of dividing an observation target area into a plurality of sections. [Figure 4] 1 is a block diagram showing the configuration of a main part of a traffic information generating device according to this embodiment. [Figure 5]FIG. 1 illustrates vehicle tracking data. [Figure 6] 10 is a flowchart showing a tracking process. [Figure 7] 10 is a flowchart showing a travel time generation process. [Figure 8] FIG. 10 is a block diagram showing the configuration of a main part of a traffic information generating device according to a first modified example. [Figure 9] 10 is a flowchart showing a signal estimation process. [Figure 10] FIG. 10 is a block diagram showing the configuration of a main part of a traffic information generating device according to a second modification. [Figure 11] FIG. 10 is a diagram showing tracking data of a traffic information generating device according to a second modification. [Figure 12] 10 is a flowchart showing a congestion length estimation process. DETAILED DESCRIPTION OF THE INVENTION

[0021] Hereinafter, an embodiment of the present invention will be described.

[0022] <1. Application Examples> 1 and 2 are schematic diagrams showing examples of observation target areas for which the traffic information generating device of this example generates traffic information. In this example, as shown in FIG. 1, the observation target area SA is set as an entrance road to an intersection. In the example shown in FIG. 1, the observation target area SA is set as one of the entrance roads to the intersection. The number of entrance roads to the intersection for which the observation target area SA is set is not limited to one, and may be multiple. In this example, the traffic information generating device generates travel times for the observation target area SA as traffic information.

[0023] The main road at the intersection shown in FIG. 1 is a road with three lanes in each direction (a side lane, a center lane, and a right-turn lane). The secondary road at the intersection shown in FIG. 1 is a road with two lanes in each direction (a driving lane and a right-turn lane). The intersection is equipped with traffic lights 100 (100a-100d) for each oncoming road. Traffic lights 100a and 100b illuminate a color signal indicating whether a driver of a vehicle 110 traveling on the main road has the right of way through the intersection. Traffic lights 100c and 100d illuminate a color signal indicating whether a driver of a vehicle 110 traveling on the secondary road has the right of way through the intersection. A signal control device (not shown) controls the switching of the color signals of traffic lights 100.

[0024] Traffic light 100a turns on a light signal indicating whether the driver of vehicle 110 traveling on the approach road where observation area SA is set has the right of way at the intersection. In addition, the light signals of traffic lights 100a and 100b are switched synchronously, and the light signals of traffic lights 100c and 100d are also switched synchronously.

[0025] As shown in FIG. 2, the radio radar 2 emits millimeter waves as search waves and receives the reflected waves to detect the position and speed of the object (vehicle 110) that reflected the search waves. The radio radar 2 scans the observation area SA with the search waves. The radio radar 2 repeats scanning the observation area SA with the search waves, for example, at intervals of 100 msec. That is, the radio radar 2 repeatedly detects vehicles 110 traveling within the observation area SA at the same period as it scans the observation area SA with the search waves. Each time the radio radar 2 scans the observation area SA with the search waves, it outputs detection data for the vehicle 110, including the position and speed of the vehicle 110 detected during the current scan.

[0026] The traffic information generating device 1 collects detection data of the vehicle 110 output by the radio radar 2 and generates tracking data of the vehicle 110 that has traveled through the observation area SA. The tracking data is data in which the position and speed of each vehicle 110 are arranged in chronological order.

[0027] The traffic information generating device 1 in this example generates section traffic information for the vehicle 110 in each section obtained by dividing the observation area SA in the traveling direction of the vehicle 110. In this example, as shown in Fig. 3, the observation area SA is divided into five sections (section 1 to section 5) at 30 m intervals. The radio radar 2 is installed on the entrance side (section 1 side) of the intersection.

[0028] In this example, the observation area SA is divided at equal intervals (at 30 m intervals in the example shown in FIG. 3) in the traveling direction of the vehicle 110, but the length of the sections may be shorter the further upstream the detection accuracy of the vehicle 110 by the radio radar 2 decreases. In other words, the observation area SA may be divided so that the length of the sections becomes shorter the farther away from the installation position of the radio radar 2. For example, the observation area SA may be divided so that the length of the first section shown in FIG. 3 is 40 m, the length of the second section is 35 m, the length of the third section is 30 m, the length of the fourth section is 25 m, and the length of the fifth section is 20 m.

[0029] In this example, the traffic information generating device 1 generates section traffic information representing the travel time of the vehicle 110 in each section obtained by dividing the observation area SA. The traffic information generating device 1 integrates the travel times generated for each section to generate the travel time of the vehicle 110 in the observation area SA (traffic information for the observation area SA).

[0030] A vehicle 110 that can be detected without interruption from entering the observation area SA to leaving it (from entering the fifth section to leaving the first section) is a vehicle 110 that can be detected without interruption in each section (first section to fifth section). Even if a vehicle 110 is undetectable in a certain section (a vehicle 110 that cannot be detected without interruption from entering the observation area SA to leaving it), it may still be detected without interruption in some other section. For example, a vehicle 110 that is undetectable in the third section (a vehicle 110 that does not miss detection from entering the fifth section until halfway through the third section) can be detected without interruption in at least the fourth and fifth sections. It is also possible that this vehicle 110 can be detected without interruption in either the first section or the second section, or both. Therefore, in each section, the number of vehicles 110 that can be detected without interruption in that section is equal to or greater than the number of vehicles 110 that can be detected without interruption from entering to leaving the observation area SA.

[0031] In addition, if there are no missed detections of vehicles 110, the number of vehicles 110 that can be detected without interruption from entering the observation area to leaving it will be the same as the number of vehicles 110 that can be detected without interruption in each section.

[0032] The accuracy of the section traffic information (section travel time in this example) for each section can be improved because the number of vehicles 110 from which detection data used to generate the information can be reduced. As described above, the traffic information generating device 1 generates traffic information for the observation area SA by integrating the section traffic information generated for each section. Therefore, the traffic information generating device 1 can improve the accuracy of the traffic information for the observation area SA that is generated based on the detection results of the vehicles 110 by the radio radar 2 (radar device).

[0033] <2.Configuration example> 4 is a block diagram showing the configuration of the main parts of the traffic information generating device 1 of this example. The traffic information generating device 1 of this example includes a control unit 11, a detection data input unit 12, a tracking database 13 (tracking DB 13), a timer unit 14, and an input / output unit 15.

[0034] The control unit 11 controls the operation of each part of the main body of the traffic information generating device 1. The control unit 11 also has a sensed data processing unit 11a, a generating unit 11b, and an integrating unit 11c. The sensed data processing unit 11a, the generating unit 11b, and the integrating unit 11c of the control unit 11 will be described later.

[0035] The radio radar 2 is connected to the detection data input unit 12. Every time the radio radar 2 scans the observation area SA with a search wave, it outputs detection data for each vehicle 110 detected in the current scan, which corresponds to the position and speed of that vehicle 110. The detection data output by the radio radar 2 is input to the detection data input unit 12.

[0036] The tracking DB 13 stores tracking data of vehicles 110 detected by the radio radar 2. FIG. 5 is a diagram showing the tracking data. The tracking data is stored for each vehicle 110. The tracking data is data in which an ID that identifies the vehicle 110 is associated with detection data at each detection time. The detection time may be, for example, the time when the radio radar 2 starts scanning the observation area SA with the search wave, the time when the radio radar 2 finishes scanning, a time between the start time and the end time, or the time when the detection data is input from the radio radar 2.

[0037] The detection data includes the speed and position of the detected vehicle 110. The position is the position in the direction of travel of the vehicle 110 in the observation area SA and the position in the width (total width) direction of the vehicle 110. The position in the direction of travel of the vehicle 110 corresponds to the distance from the installation position of the radio radar 2 to the vehicle 110. Furthermore, the position in the width direction of the vehicle 110 corresponds to the lane in which the vehicle 110 is traveling.

[0038] The timekeeping unit 14 is a timer that keeps track of the current time.

[0039] The input / output unit 15 inputs and outputs data to and from a higher-level device (for example, a server device) installed in, for example, a control center.

[0040] Next, the sensed data processing unit 11a, the generating unit 11b, and the integrating unit 11c included in the control unit 11 will be described.

[0041] The detection data processing unit 11a processes the detection data of the vehicle 110 input from the radio radar 2 and stores it in the tracking DB 13. The detection data processing unit 11a performs an identification process to associate the vehicle 110 detected by the radio radar 2 during the current scan of the observation area SA with the vehicle 110 detected during the previous scan of the observation area SA with the search waves. The detection data processing unit 11a sets the ID of the vehicle 110 detected by the radio radar 2 during the current scan of the observation area SA with the search waves to the ID of the associated vehicle 110 detected during the previous scan of the observation area SA with the search waves. Furthermore, if the detection data processing unit 11a cannot associate the vehicle 110 detected by the radio radar 2 during the current scan of the observation area SA with the vehicle 110 detected during the previous scan of the observation area SA with the search waves, the detection data processing unit 11a assigns a new ID to the currently detected vehicle 110.

[0042] Note that there are cases where the vehicles 110 detected in the previous scan of the observation area SA with the probe wave cannot be matched with the vehicles 110 detected in the current scan of the observation area SA with the probe wave. For example, the vehicles 110 detected in the previous scan of the observation area SA with the probe wave may have already moved outside the observation area SA when the current scan of the observation area SA with the probe wave is performed, or may be vehicles 110 that were not detected this time.

[0043] The generation unit 11b determines whether it is time to generate traffic information. If it is determined that it is time to generate traffic information, the generation unit 11b generates section travel times as section traffic information for each section into which the observation target area SA is divided. The generation time for traffic information is, for example, a time when a preset time (for example, 5 to 10 minutes) has elapsed since it was determined that it was time to generate the previous traffic information.

[0044] When the generation unit 11b determines that it is time to generate traffic information, it extracts, for each section into which the observation target area SA is divided, vehicles 110 to be used in generating section traffic information for that section. The vehicles 110 extracted here are vehicles 110 that have exited that section during the period from the time when the previous traffic information was generated to the time when the current traffic information is generated, and that have been detected continuously in that section.

[0045] The generation unit 11b calculates the section travel time, which is section traffic information, for each section using the detection data of the extracted vehicles 110. For example, the generation unit 11b extracts the section travel time for the fifth section by extracting vehicles 110 that passed the boundary between the fifth section and the fourth section (e.g., the 120-m line shown in FIG. 3) and were detected continuously in the fifth section (e.g., the period from passing the 150-m line to passing the 120-m line shown in FIG. 3) between the timing of generating the previous traffic information and the timing of generating the current traffic information. For each extracted vehicle 110, the generation unit 11b obtains the time required to travel the fifth section (e.g., the time difference from passing the 150-m line to passing the 120-m line shown in FIG. 3) from the tracking data shown in FIG. 3. This time required to travel the fifth section corresponds to the tracking time referred to in this invention. The generating unit 11b generates the average value of the time required to travel the fifth section acquired for each vehicle 110 as the section travel time for this fifth section.

[0046] For the first to fourth sections, the section travel times for those sections are generated in the same manner.

[0047] Note that, here, for each section, the section travel time for that section is the average value of the time required for the extracted vehicle 110 to travel that section, but it may also be the median, mode, root mean square, etc. of the time required for the extracted vehicle 110 to travel that section. Also, the section travel time may be calculated by calculating the average speed of the extracted vehicle 110 in that section and dividing the length of that section by the average speed.

[0048] The integrating unit 11c integrates the travel times of the sections generated by the generating unit 11b to generate the travel time for the observation area SA. For example, the integrating unit 11c generates the sum of the section travel times for the sections as the travel time for the observation area SA. Specifically, the integrating unit 11c Travel time in observation area SA = (sum of section travel times for each section) The result of the calculation is used as the travel time for the observation area SA. Furthermore, the integrating unit 11c may take into consideration waiting at traffic lights at intersections and perform a calculation to generate, for example, a value obtained by multiplying the sum of the section travel times of each section by a predetermined constant α (α>1) as the travel time for the observation area SA. Travel time in observation area SA = (total travel time for each section) × α The result of the calculation is used as the travel time for the observation area SA. The integrating unit 11c may also perform a calculation to generate the travel time for the observation area SA by adding the sum of the section travel times for each section to the value obtained by multiplying the red light time Tr of the traffic light 100a by a proportional constant β. Travel time in observation area SA = (total travel time for each section) + Tr × β In this case, β may be set based on, for example, the ratio between the number of vehicles 110 that stop within the observation area SA due to the red light of traffic light 100a (vehicles 110 that are affected by the red light of traffic light 100a) and the number of vehicles 110 that enter the intersection within the observation area SA without stopping due to the red light of traffic light 100a (vehicles 110 that are not affected by the red light of traffic light 100a).

[0049] The method by which the integrating unit 11c integrates the section travel times of the sections when generating the travel time for the observation target area SA is not limited to the above-described method, and other methods may also be used.

[0050] The control unit 11 of the traffic information generating device 1 is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the traffic information generating program according to the present invention, it operates as a detected data processing unit 11a, a generating unit 11b, and an integrating unit 11c. The memory also has an area for expanding the traffic information generating program according to the present invention and an area for temporarily storing data generated during execution of the traffic information generating program. The control unit 11 may be an LSI that integrates the hardware CPU, memory, and the like. The hardware CPU is also a computer that executes the traffic volume measurement method according to the present invention.

[0051] <3. Example of operation> The traffic information generating device 1 in this example performs the following tracking process and travel time generating process.

[0052] The tracking process is a process of processing the detection data of the vehicle 110 input from the radio radar 2, generating the tracking data shown in Fig. 5, and storing it in the tracking DB 13. Fig. 6 is a flowchart showing this tracking process.

[0053] The traffic information generating device 1 waits (s1) for the detection data output by the radio radar 2 to be input to the detection data input unit 12. In this example, the radio radar 2 scans the observation area SA with a search wave, and outputs detection data for each vehicle 110 detected during this scan, associating the position and speed of that vehicle 110. The radio radar 2 also scans the observation area SA with a search wave at 100 msec intervals. Therefore, in this example, the detection data is input to the detection data input unit 12 at 100 msec intervals.

[0054] The detection data processing unit 11a selects a target vehicle from among the vehicles 110 detected this time that have not been subjected to the processes in s3 to s7 described below (unprocessed vehicles 110) (s2). The detection data processing unit 11a determines whether the target vehicle selected in s2 is a vehicle 110 that was detected previously (s3). In s3, the position and speed of the target vehicle are used to determine whether the corresponding vehicle 110 was detected previously.

[0055] If the target vehicle is a vehicle 110 that was also detected last time, the detection data processing unit 11a matches the ID of this target vehicle to the ID assigned last time (s4), and adds the detection data of the current target vehicle to the tracking data of the vehicle 110 with this ID stored in the tracking DB 13 (s5).

[0056] If the target vehicle is a vehicle 110 that was not detected last time, the detection data processing unit 11a issues and assigns a new ID to this target vehicle (s6). The detection data processing unit 11a registers the tracking data of the target vehicle to which the newly issued ID has been assigned in the tracking DB 13 (s7). In s6, an ID that can be distinguished from other vehicles 110 is issued. At this point, the detection data included in the tracking data registered in the tracking DB 13 in s7 is only the current detection data.

[0057] After performing the process of s5 or s7, the detection data processing unit 11a determines whether or not there is an unprocessed vehicle 110 that has been detected this time by the radio radar 2 and has not yet been subjected to the processes of s3 to s7 (s8). If there is an unprocessed vehicle 110, the detection data processing unit 11a returns to s2. If there is no unprocessed vehicle 110, the detection data processing unit 11a returns to s1.

[0058] The traffic information generating device 1 of this example executes this tracking process to obtain tracking data indicating the travel trajectory of the vehicle 110 detected by the radio radar 2 within the observation area SA.

[0059] As is clear from the above explanation, the tracking data of each vehicle 110 registered in the tracking DB 13 is based on the detection result of the vehicle 110 by the radio radar 2.

[0060] Next, a description will be given of the travel time generation process of the traffic information generation device 1. Fig. 7 is a flowchart showing this travel time generation process. In this example, the traffic information generation device 1 generates travel times for the observation area SA as traffic information for this observation area SA.

[0061] The generation unit 11b waits for the timing to generate travel times for the observation target area SA (s11). The timing to generate traffic information is, for example, the timing when a preset time (for example, 5 to 10 minutes) has elapsed since it was determined that it was the timing to generate the previous traffic information.

[0062] When the generation unit 11b determines that it is time to generate traffic information, it selects one of the sections into which the observation target area SA is divided (five sections, Section 1 to Section 5, in the example shown in FIG. 3), which has not undergone the processes in s13 and s14 described below (unprocessed section), as a target section (s12). The generation unit 11b extracts vehicles 110 that have exited the selected target section during the period from the time when the previous traffic information was generated to the time when the current traffic information is generated, and that have been detected continuously in the target section by the radio radar 2 (s13). The vehicles 110 extracted in s13 are vehicles 110 whose travel trajectories from entering to leaving the target section are registered in the tracking DB 13.

[0063] The generation unit 11b calculates, for each vehicle 110 extracted in s13, the time required for that vehicle 110 to travel through the target section (i.e., the time difference between the time when the vehicle 110 exits the target section and the time when the vehicle enters the target section). The generation unit 11b calculates the section travel time for this target section based on the time required for travel through the target section calculated for each vehicle 110 (s14). In s14, the generation unit 11b calculates the average value of the time required for travel through the target section calculated for each vehicle 110 as the section travel time for this target section.

[0064] As described above, the generation unit 11b may calculate, as the section travel time of this target section, the median, mode, root mean square, or the like of the time required to travel the target section calculated for each vehicle 110. Furthermore, the generation unit 11b may calculate the average speed of the target section for the extracted vehicle 110, and calculate the section travel time by dividing the length of the section by the average speed.

[0065] The generation unit 11b determines whether there are any sections (unprocessed sections) for which the processes in s13 and s14 have not been performed (s15), and if there are any unprocessed sections, the process returns to s12. If the generation unit 11b determines that there are no unprocessed sections, the integration unit 11c integrates the section travel times of each section, calculates the travel time for the observation target area SA (s16), and the process returns to s11.

[0066] For example, the integrating unit 11c generates the sum of the section travel times of the sections as the travel time for the observation area SA. Travel time in observation area SA = (sum of section travel times for each section) The result of the calculation is used as the travel time for the observation area SA. Furthermore, the integrating unit 11c may take into consideration waiting at traffic lights at intersections and perform a calculation to generate, for example, a value obtained by multiplying the sum of the section travel times of each section by a predetermined constant α (α>1) as the travel time for the observation area SA. Travel time in observation area SA = (total travel time for each section) × α The result of the calculation is used as the travel time for the observation area SA. The integrating unit 11c may also perform a calculation to generate the travel time for the observation area SA by adding the sum of the section travel times for each section to the value obtained by multiplying the red light time Tr of the traffic light 100a by a proportional constant β. Travel time in observation area SA = (total travel time for each section) + Tr × β In this case, β may be set based on, for example, the ratio between the number of vehicles 110 that stop within the observation area SA due to the red light of traffic light 100a (vehicles 110 that are affected by the red light of traffic light 100a) and the number of vehicles 110 that enter the intersection within the observation area SA without stopping due to the red light of traffic light 100a (vehicles 110 that are not affected by the red light of traffic light 100a).

[0067] The method by which the integrating unit 11c integrates the section travel times of the sections when generating the travel time for the observation target area SA is not limited to the above-described method, and other methods may also be used.

[0068] The red signal duration Tr of the traffic light 100a may be obtained in advance from signal control parameters used by the traffic light control device to control the red signal of the traffic light 100, or may be measured using a sensor. For example, the duration Tr may be measured by using a current sensor to detect the current flowing through a power supply line that supplies power to turn on the red signal of the traffic light 100a, or by using an image sensor to detect and measure the on / off of the red signal of the traffic light 100a, or by other methods of detection and measurement.

[0069] In this way, the traffic information generating device 1 of this example calculates the section travel time for each section obtained by dividing the observation target area SA in the traveling direction of the vehicle 110, and integrates the section travel times for each section to calculate the travel time for the observation target area SA. Since a decrease in the detection data of the vehicle 110 used to generate the section travel time for each section is suppressed, the section travel time can be generated with high accuracy. The traffic information generating device 1 of this example integrates the section travel times generated for each section to generate the travel time for the observation target area SA. Therefore, the traffic information generating device 1 can improve the accuracy of the travel time for the observation target area SA.

[0070] The traffic information generating device 1 in the above example has been described as generating the travel time in the observation area SA as the traffic information for the observation area SA, but may instead generate, for example, the average speed of the vehicle 110 in the observation area SA as the traffic information. Furthermore, the traffic information generating device 1 may generate both the travel time of the vehicle 110 in the observation area SA and the average speed of the vehicle 110 in the observation area SA as the traffic information.

[0071] The traffic information generating device 1 also outputs the generated traffic information for the observation target area SA to a higher-level device connected to the input / output unit 15.

[0072] The traffic information generating device 1 may generate traffic information for each lane in the observation area SA.

[0073] <4. Modifications> Variation 1 The traffic information generating device 1A of this modified example 1 performs the same processing as the traffic information generating device 1 of the above example, and also performs processing to estimate the green light start time and the red light start time of the traffic light 100a. By performing this processing, the traffic information generating device 1A can estimate the red light duration of the traffic light 100a (the time from when the red light starts to when the green light starts).

[0074] Fig. 8 is a diagram showing the configuration of the main parts of a traffic information generating device of Modification 1. The traffic information generating device 1A of Modification 1 differs from the above-described traffic information generating device 1 in that a control unit 11A additionally has a signal state estimating unit 11d. In Fig. 8, the same components as those in Fig. 4 are denoted by the same reference numerals.

[0075] The signal state estimation unit 11d estimates the start time of the green light and the start time of the red light of the traffic light 100a. The signal state estimation unit 11d estimates the start time of the green light and the start time of the red light of the traffic light 100a for each lane based on the detection results of vehicles 110 passing through a position relatively close to the intersection and downstream of the observation target area SA. For example, the signal state estimation unit 11d estimates the start time of the green light and the start time of the red light of the traffic light 100a for each lane based on the detection results of vehicles 110 passing through the approach line shown in FIG. 3. The signal state estimation unit 11d sets the earliest of the start times of the green light of the traffic light 100a estimated for each lane as the start time of the green light of the traffic light 100a. Furthermore, the signal state estimation unit 11d sets the latest of the start times of the red light of the traffic light 100a estimated for each lane as the start time of the red light of the traffic light 100a.

[0076] 9 is a flowchart showing the signal estimation process performed by the signal state estimation unit 11d to estimate the green light start time and the red light start time of the traffic light 100a.

[0077] Here, the traffic information generating device 1A will be described assuming that the current light color of the traffic light 100a is red.

[0078] The signal state estimation unit 11d determines for each lane whether the number of vehicles 110 that have passed through the point where the approach line is set has reached the green start number (e.g., 3) during the most recent predetermined first time period (e.g., 5 seconds) (s31). The signal state estimation unit 11d first sets the lane for which the number of vehicles 110 that have passed through the point where the approach line is set has reached the green start number during the most recent predetermined first time period as the lane to be determined for the green start time.

[0079] The signal state estimation unit 11d estimates the tentative green start time as the time of passage of the vehicle 110 that first passed the point where the approach line is set in the most recent second hour in the lane that is the target of determination of the green start time (s32). The tentative green start time estimated in s32 is a time that is at most one hour in the past from the current time (the time when the number of vehicles 110 that passed the point where the approach line is set reaches the green start number).

[0080] The signal state estimation unit 11d estimates the green start time of the traffic light 100a to be a time that is a predetermined estimated correction time (e.g., 3 seconds) before the tentative green start time (s33). This estimated correction time is determined taking into consideration the distance from the entrance to the intersection to the point where the entry line is set, a departure delay of the vehicle 110, etc.

[0081] After estimating the start time of the green light in s33, the signal state estimation unit 11d determines, for each lane, whether or not a vehicle 110 has passed through the point where the entry line is set for the most recent predetermined second time period (10 seconds in this example) (s34). If the signal state estimation unit 11d determines that a vehicle 110 has not passed through any lane for the most recent second time period, it designates that lane as a non-target lane. If there is a lane (target lane) that has not been determined as a non-target lane (target lane) (s35), the signal state estimation unit 11d returns to s34.

[0082] When the signal state estimation unit 11d determines that all lanes are non-target lanes, it estimates the time when the passage of a vehicle 110 was first not detected in the lane last determined to be a non-target lane over the most recent two hours (i.e., two hours before the present time) as the tentative red light start time (s36).

[0083] The signal state estimation unit 11d estimates the time that is a predetermined estimated correction time (for example, 3 seconds) before the tentative red light start time as the red light start time (s37), and returns to s31.

[0084] In this way, the traffic information generating device 1A of this modified example 1 can estimate the start time of the green light and the start time of the red light of the traffic light 100a based on the detection result of the vehicle 110 that has passed through the approach line by the radio radar 2. Therefore, because the traffic information generating device 1A of this modified example 1 detects the state of the light signal of the traffic light 100a based on the detection result of the vehicle 110 by the radio radar 2, there is no need to install the above-mentioned current sensor, image sensor, etc., and the cost required for building the system can be reduced.

[0085] In addition, in the above-described first modification, the cycle length of traffic light 100 may be estimated, and the green start time and red start time may be estimated based on the estimated cycle length. In this way, it is possible to reduce delays in estimating the green start time and red start time.

[0086] For example, the cycle length of a traffic light is Cycle Length = Previous cycle length × γ + (current green start time – previous green start time) × (1 – γ) where γ is a smoothing coefficient and satisfies 0<γ<1.

[0087] By estimating the cycle length in this manner, the next green start time (the time when the cycle length has elapsed since the estimated green start time) can be estimated at the time when the green start time is estimated in the above-mentioned process. Similarly, the next red start time (the time when the cycle length has elapsed since the estimated red start time) can be estimated at the time when the red start time is estimated.

[0088] The cycle length may be updated each time the green start time or red start time is estimated.

[0089] Variation 2 FIG. 10 is a diagram showing the configuration of the main part of the traffic information generating device of the second modification.

[0090] As shown in FIG. 10, the traffic information generating device 1B of the second modification differs from the traffic information generating device 1A of the first modification in that a control unit 11B includes a vehicle state determining unit 11e and a congestion length estimating unit 11f.

[0091] The vehicle state determination unit 11e determines the state of the vehicle 110 detected by the radio radar 2. In this example, the state of the vehicle 110 refers to six states: running, accelerating, a starting event, decelerating, a stopping event, and stopped.

[0092] Accelerating means that the previous state was running and the acceleration was a predetermined value (for example, 0.2 m / sec 2 ) or more. A starting event is a state in which a predetermined starting speed (for example, 5 m / sec) is reached from a stopped state. Deceleration means that the previous state was driving and the deceleration was a predetermined value (for example, 0.2 m / sec 2 ) or more. A stop event is a state in which the previous state was deceleration and the speed was equal to or less than a predetermined stop speed (for example, 3 m / sec). "Stopped" refers to a state in which the previous event was a stop event or the vehicle is currently stopped and the speed is equal to or less than a predetermined stop speed (for example, 3 m / sec). The term "driving" refers to a state other than the above-mentioned accelerating, starting event, decelerating, stopping event, and stopped state.

[0093] The vehicle state determination unit 11e determines the state of the vehicle 110, for example, every second. That is, the state of the vehicle 110 is updated every second. For each vehicle 110 detected by the radio radar 2, the vehicle state determination unit 11e determines the state by referring to the tracking data of the vehicle 110 registered in the tracking DB 13. Furthermore, in this modification 2, when the vehicle state determination unit 11e determines the state of each vehicle 110, it registers the state determination result in the tracking data of the vehicle 110. That is, in this modification 2, the state of the vehicle 110 is associated with the detection data, as shown in FIG. 11 .

[0094] Note that the detection data that is not associated with a state is the detection data that was not used to determine the state. In this example, the state of the vehicle 110 is determined at intervals of 1 second, and the detection of the vehicle 110 by the radio radar 2 is performed at intervals of 100 msec.

[0095] Furthermore, the vehicle state determination unit 11e of this modified example 2 does not determine that the vehicle 110 that was previously determined to be a start event is a start event again. Furthermore, the vehicle state determination unit 11e does not determine that the vehicle 110 that was previously determined to be a stop event is a stop event again.

[0096] The traffic information generating device 1B of this second modification can determine the state of the vehicle 110 detected by the radio radar 2 every second, and can obtain more detailed information about the traffic conditions in the observation area SA.

[0097] The congestion length estimation unit 11f also calculates the congestion length of each lane using the state of the vehicle 110 determined by the vehicle state determination unit 11e.

[0098] 12 is a flowchart showing the traffic congestion length estimation process. The traffic congestion length estimation unit 11f waits for the vehicle state determination unit 11e to determine the state of the vehicle 110 (s41). Once the state of the vehicle 110 has been determined, the traffic congestion length estimation unit 11f determines a target lane for estimating the traffic congestion length (s42), and extracts vehicles 110 located in this target lane that have been determined to be involved in the current stop event (s43). The traffic congestion length estimation unit 11f estimates the extracted last vehicle 110 that has been involved in the stop event as the last vehicle 110 of the traffic congestion. The traffic congestion length estimation unit 11f calculates the traffic congestion length based on the currently estimated position (stop wave position) of the last vehicle 110 of the traffic congestion (s44). The traffic congestion length is, for example, the distance from the installation position of the radio radar 2 to the currently estimated position of the last vehicle 110 of the traffic congestion.

[0099] The congestion length estimation unit 11f determines whether or not there are any unprocessed lanes (s45), and if there are any unprocessed lanes, the process returns to s42 and repeats the above processing. If there are no unprocessed lanes, the congestion length estimation unit 11f returns to s41 and repeats the above processing.

[0100] In this way, the congestion length estimation unit 11f estimates the congestion length of each lane at each cycle (every second in this example) when the vehicle state determination unit 11e determines the state of the vehicle 110.

[0101] Furthermore, when the congestion length estimation unit 11f determines that the position of the vehicle 110 that was the currently extracted stop event is inappropriate relative to the position of the last vehicle 110 in the previously estimated congestion length, the congestion length estimation unit 11f may be configured not to use the position of the vehicle 110 that was determined to be the stop event for estimating the congestion length. For example, the congestion length estimation unit 11f determines that the congestion length is inappropriate when the stop event of the vehicle 110 is such that the congestion length calculated this time is shorter than the congestion length calculated previously by a predetermined percentage (for example, 90%) or more, and when the extension speed of the congestion length calculated this time is greater than the congestion length calculated previously by a predetermined speed (for example, 15 m / sec) or more.

[0102] This configuration can prevent the congestion length estimated by the congestion length estimation unit 11f from becoming inappropriate, that is, it can prevent a decrease in the accuracy of the congestion length estimation by the congestion length estimation unit 11f.

[0103] In this way, the traffic information generating device 1B of the second modification can obtain the congestion length of each lane within the observation target area SA.

[0104] Furthermore, as the length of the traffic jam increases, the detection accuracy of the vehicle 110 by the radio radar 2 decreases, and there is a possibility that a vehicle 110 in a stop event will not be detected. In this case, the traffic jam length estimation unit 11f may calculate the traffic jam length (the position of the last vehicle 110 in the traffic jam) using the estimated traffic volume within the observation target area SA. For example, if the time at which the latest stop event was detected is between the start time of the red light estimated by the signal state estimation unit 11d and the start time of the subsequent green light, Position of car 110 at the end of the traffic jam = Position of vehicle 110 that detected the latest stop event + Elapsed time since the latest stop event was detected × Exponentially smoothed traffic volume at the upstream point × Average headway Furthermore, for example, if the time at which the latest stop event was detected is between the start time of the green light estimated by the signal state estimation unit 11d and the start time of the subsequent red light, Position of the last vehicle in the traffic jam (stop wave position) = Elapsed time since the previous red light start time × exponentially smoothed traffic volume at upstream point / estimated period of exponentially smoothed traffic volume at upstream point × average headway It is calculated as follows.

[0105] The upstream point exponentially smoothed traffic volume estimation period is, for example, 2.5 minutes (150 seconds). The average headway is, for example, 7 m. The upstream point exponentially smoothed traffic volume is calculated based on the number of vehicles 110 that have passed the 120 m line or the 150 m line shown in FIG. 3.

[0106] Specifically, for each lane, the exponentially smoothed traffic volume at the upstream point is Exponential smoothing traffic volume at upstream points = Previous exponentially smoothed traffic volume at the upstream point × (1-δ) + Number of detected vehicles at the upstream point × δ where δ is a smoothing coefficient and is in the range of 0<δ<1. The number of vehicles detected at the upstream point is the number of vehicles 110 that passed the upstream point detected by the radio radar 2 during the period from the previous estimation timing to the current estimation timing.

[0107] In this way, the congestion length estimation unit 11f can accurately estimate the congestion length even if a stop event of the vehicle 110 is missed. In other words, the position of the stop wave of the vehicle 110 within the observation target area SA can be accurately estimated.

[0108] The congestion length estimation unit 11f also estimates the starting wave position. Specifically, the congestion length estimation unit 11f estimates the starting wave position as follows: Starting wave position = time elapsed since the start of blue light × starting wave speed The starting wave speed may be set to, for example, 5 m / sec.

[0109] As a result, the traffic information generating device 1B of this variant example 2 can obtain the sections where each vehicle 110 stopped within the observation area SA and the stopping time in the stopped section (the time difference between the departure time (arrival time of the departure wave) and the stopping time (arrival time of the stopping wave)).

[0110] The traffic information generating device 1B can estimate travel time for vehicles 110 for which only the entry time or exit time into a section is measured, by applying the stop time estimated from the stop wave position and departure wave position, and by increasing the number of vehicles 110 to be measured, the travel time in the observation area SA can be calculated with even greater accuracy.

[0111] The traffic information generating device 1B calculates the sum of the section travel times calculated for each section as the travel time for the observation area SA. Therefore, the traffic information generating device 1B can calculate the travel time for the observation area SA with even greater accuracy.

[0112] In addition, the radio radar 2 in the above example may be replaced with a laser radar. The cycle for calculating travel time within the observation area SA may be set to coincide with the cycle of the traffic lights 100, or may be set at appropriate time intervals.

[0113] It should be noted that this invention is not limited to the above-described embodiments, and that the components can be modified and embodied in practice without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, some components may be omitted from all the components shown in the embodiments. Furthermore, components from different embodiments may be appropriately combined. Furthermore, the order of each step in the flowcharts described in all the above examples is merely an example, and may be appropriately changed within the scope of the present invention.

[0114] Furthermore, the correspondence between the configuration according to the present invention and the configuration according to the above-described embodiment can be described as follows: <Additional Notes> a detection data acquisition unit (11a) that acquires detection data of a vehicle (110) traveling in an observation area (SA) detected by a radar device (2) scanning the observation area (SA) with a search wave; a generation unit (11b) that processes the detection data of the vehicle (110) acquired by the detection data acquisition unit (11a) and generates section traffic information that indicates the state of traffic flow of the vehicle (110) in each section, the section being obtained by dividing the observation target area (SA) in the traveling direction of the vehicle (110); A traffic information generating device (1) comprising an integration unit (11c) that integrates the section traffic information generated for each section by the generation unit (11b) and generates traffic information for the observation target area (SA). [Explanation of symbols]

[0115] 1, 1A, 1B…Traffic information generation device 2...Radio wave radar 11, 11A, 11B...Control unit 11a...Detection data processing unit 11b...Generation section 11c...Integration Department 11d...Signal state estimation unit 11e...Vehicle state determination unit 11f...Congestion length estimation section 12...Detection data input section 13...Tracking database (Tracking DB) 14...Timekeeping section 15…Input / output section 100(100a~100d)…Traffic light 110...Vehicle

Claims

1. a detection data acquisition unit that acquires detection data of vehicles traveling in an observation target area detected by a radar device by scanning the observation target area with a search wave; a generation unit that processes the vehicle detection data acquired by the detection data acquisition unit and generates section traffic information that indicates the state of vehicle traffic flow in each section, for each of a plurality of sections that are obtained by dividing the observation target area in the vehicle traveling direction; an integration unit that integrates the section traffic information generated for each section by the generation unit and generates traffic information for the observation target area, the generation unit generates, for each section obtained by dividing the observation target area in a vehicle traveling direction, a travel time for that section as the section traffic information; the integrating unit calculates a travel time for the observation target area based on a sum of the travel times generated for each section by the generating unit, and generates the calculated travel time for the observation target area as the traffic information for the observation target area. Traffic information generation device.

2. 2. The traffic information generating device according to claim 1, wherein the generating unit extracts vehicles that can be tracked from a start point to an end point of each section obtained by dividing the observation area in the vehicle travel direction, and generates a travel time for the section using the tracking time for which each extracted vehicle was detected in the section.

3. 3. The traffic information generating device according to claim 2, wherein the generating unit generates, for each section obtained by dividing the observation area in the vehicle's traveling direction, a travel time for the section using the tracking time and a vehicle stopping state in the section detected based on the vehicle detection data acquired by the detection data acquiring unit.

4. A detection data acquisition unit that acquires detection data of vehicles traveling in an observation area detected by a radar device by scanning the observation area with a search wave; a generation unit that processes the vehicle detection data acquired by the detection data acquisition unit and generates section traffic information that indicates the state of vehicle traffic flow in each section, for each of a plurality of sections that are obtained by dividing the observation target area in the vehicle traveling direction; an integration unit that integrates the section traffic information generated for each section by the generation unit and generates traffic information for the observation target area, the generation unit extracts vehicles that can be tracked from the start point to the end point of each section obtained by dividing the observation area in the vehicle travel direction, and generates the travel time for each section as the section traffic information using the tracking time for each extracted vehicle that was detected in that section and the stopping state of the vehicle in that section detected based on the vehicle detection data acquired by the detection data acquisition unit; The integrating unit generates a travel time for the observation target area as the traffic information for the observation target area. Traffic information generation device.

5. a detection data acquisition step of acquiring detection data of vehicles traveling in an observation area detected by the radar device by scanning the observation area with a search wave; a generation step of processing the vehicle detection data acquired in the detection data acquisition step, and generating section traffic information indicating the state of vehicle traffic flow in each of a plurality of sections obtained by dividing the observation target area in the vehicle traveling direction; an integration step of integrating the section traffic information generated for each section in the generation step to generate traffic information for the observation target area, the generating step is a step of generating, for each section obtained by dividing the observation target area in a vehicle traveling direction, a travel time for the section as the section traffic information; A traffic information generation method, wherein the integration step is a step of calculating the travel time for the observation area based on the sum of the travel times generated for each section in the generation step, and generating the calculated travel time for the observation area as the traffic information for the observation area.

6. A detection data acquisition step of acquiring detection data of vehicles traveling in an observation area detected by a radar device by scanning the observation area with a search wave; a generation step of processing the vehicle detection data acquired in the detection data acquisition step, and generating section traffic information indicating the state of vehicle traffic flow in each of a plurality of sections obtained by dividing the observation target area in the vehicle traveling direction; an integration step of integrating the section traffic information generated for each section in the generation step to generate traffic information for the observation target area, the generating step is a step of extracting vehicles that can be tracked from the start point to the end point of each section obtained by dividing the observation area in the vehicle travel direction, and generating the travel time for each extracted vehicle as the section traffic information using the tracking time of the vehicle that was detected in that section and the stopping state of the vehicle in that section detected based on the vehicle detection data acquired in the detection data acquiring step; The traffic information generating method, wherein the integrating step is a step of generating travel times for the observation target area as the traffic information for the observation target area.

7. a detection data acquisition step of acquiring detection data of vehicles traveling in an observation area detected by the radar device by scanning the observation area with a search wave; a generation step of processing the vehicle detection data acquired in the detection data acquisition step, and generating section traffic information indicating the state of vehicle traffic flow in each of a plurality of sections obtained by dividing the observation target area in the vehicle traveling direction; an integration step of integrating the section traffic information generated for each section in the generation step to generate traffic information for the observation target area; the generating step is a step of generating, for each section obtained by dividing the observation target area in a vehicle traveling direction, a travel time for the section as the section traffic information; A traffic information generation program, wherein the integration step is a step of calculating the travel time for the observation area based on the sum of the travel times generated for each section in the generation step, and generating the calculated travel time for the observation area as the traffic information for the observation area.

8. A detection data acquisition step of acquiring detection data of vehicles traveling in an observation area detected by a radar device by scanning the observation area with a search wave; a generation step of processing the vehicle detection data acquired in the detection data acquisition step, and generating section traffic information indicating the state of vehicle traffic flow in each of a plurality of sections obtained by dividing the observation target area in the vehicle traveling direction; an integration step of integrating the section traffic information generated for each section in the generation step to generate traffic information for the observation target area; the generating step is a step of generating, for each section obtained by dividing the observation target area in a vehicle traveling direction, a travel time for the section as the section traffic information; A traffic information generation program, wherein the integration step is a step of calculating the travel time for the observation area based on the sum of the travel times generated for each section in the generation step, and generating the calculated travel time for the observation area as the traffic information for the observation area.

Images