Near miss determination system and near miss determination method

Abstract

[Problem] To improve the accuracy of collision determination and appropriately determine a near miss in collision determination that incorporates braking time, by setting an appropriate braking time for all relevant moving bodies and setting an appropriate braking time for a discretionary moving body of interest in accordance with the situation. [Solution] This roadside unit: detects a plurality of moving bodies that are present on a road around the roadside unit, which is installed on a road or in the vicinity thereof, on the basis of a detection result from a sensor provided to the roadside unit and acquires target information relating to the moving bodies (ST101); acquires information relating to a road surface state as state information relating to the road on the basis of the detection result from the sensor (ST104); sets a braking time relating to a discretionary moving body of interest on the basis of the target information and the information relating to the road surface state (ST106); and determines the risk of collision between the moving bodies on the basis of the braking time (ST107). A server determines a near miss on the basis of the determination result from the roadside unit.

Description

Near miss judgment system and near miss judgment method

[0001] The present disclosure relates to a near-miss detection system that detects near-misses involving multiple moving objects on a road, and a near-miss detection method in which processing related to the detection of near-misses is executed by a processor.

[0002] In recent years, from the perspective of traffic accident prevention, attention has been focused on near misses, which are events in which an abnormal condition with a high risk of an accident occurred but did not result in an accident. Collecting such near miss cases can provide knowledge for preventing the recurrence of near misses and reducing accidents. Therefore, a technology that can accurately and efficiently identify near misses is desired so that a large number of near miss cases can be collected.

[0003] Here, a near miss is an event in which a high risk of collision occurred but did not result in an accident, and determining a near miss first requires a collision judgment (collision prediction) regarding whether a high risk of collision occurred. In particular, the accuracy of collision judgment can be improved by focusing on the time (braking time) from when a moving object on a road, such as a vehicle, pedestrian, or bicycle, starts braking (brake operation) until it decelerates and stops.

[0004] A known technology for collision detection that takes braking time into account is one in which a device installed in a vehicle calculates the braking time of the vehicle based on the detection results of a sensor, and then performs collision detection based on that braking time and information (position and speed) of surrounding vehicles (see Patent Document 1).

[0005] Patent No. 6704531

[0006] In conventional technology, only the braking time of a mobile object equipped with a collision detection device is considered, and the braking time of a mobile object that will collide with the other mobile object is not considered. However, which mobile object's braking time should be considered depends on the situation. For example, in a typical near-miss case, when two roads connecting to an intersection are divided into priority and non-priority, a mobile object traveling on the non-priority road enters the intersection before a mobile object traveling on the priority road, and the mobile object on the priority road suddenly brakes. This presumably results in a near-miss for the driver of the mobile object on the priority road. In this case, the braking time of the mobile object on the priority road is important. For this reason, it is desirable to be able to set appropriate braking times for all related mobile objects individually, or to be able to set appropriate braking times for any mobile object of interest depending on the situation.

[0007] Therefore, the main purpose of the present disclosure is to provide a near-miss detection system and a near-miss detection method that can improve the accuracy of collision detection and appropriately detect near-misses by setting appropriate braking times for all involved moving bodies in collision detection that takes braking times into account, or by setting appropriate braking times for any moving body of interest depending on the situation.

[0008] The near-miss detection system disclosed herein is a near-miss detection system that detects near-misses involving multiple moving objects on a road, and includes a roadside unit installed on or near the road, a sensor provided in the roadside unit, and one or more processors that perform processing related to determining the near-miss based on the detection results of the sensor, wherein the processor detects multiple moving objects present on the road around the roadside unit based on the detection results of the sensor, obtains target information related to the moving objects, sets a braking time for any moving object of interest based on the target information and the detection results of the sensor, determines the risk of collision between the moving objects based on the braking time, and determines the near-miss based on the determination results.

[0009] In addition, the near-miss determination method disclosed herein is a near-miss determination method in which processing related to determining near-misses involving multiple moving objects on a road is executed by one or more processors, and is configured to detect multiple moving objects on the road around the roadside unit based on the detection results of a sensor provided in a roadside unit installed on or near the road, obtain target information related to the moving objects, set a braking time for any moving object of interest based on the target information and the detection results of the sensor, determine the risk of collision between the moving objects based on the braking time, and determine the near-miss based on the determination results.

[0010] According to the present disclosure, a braking time is set for any moving object taking into consideration road conditions, the attributes of the moving object, and the state of the moving object. This allows appropriate braking times to be set for all related moving objects, or for any moving object of interest depending on the situation, thereby improving the accuracy of collision detection and enabling appropriate near miss detection.

[0011] 1 is a block diagram showing an overall configuration of a near-miss detection system according to a first embodiment; FIG. 2 is an explanatory diagram showing an overview of the near-miss detection performed by the server according to the first embodiment; FIG. 3 is a block diagram showing a schematic configuration of a roadside device, a server, and a mobile terminal according to the first embodiment; FIG. 4 is a block diagram showing an overview of the processing performed by the roadside device and the server according to the first embodiment; FIG. 5 is a flow diagram showing a first example of the procedure of the processing performed by the roadside device according to the first embodiment; FIG. 6 is a flow diagram showing a second example of the procedure of the processing performed by the roadside device according to the first embodiment; FIG. 7 is a block diagram showing an overview of the processing performed by the roadside device and the server according to the second embodiment;

[0012] The first invention made to solve the above problem is a near-miss detection system that detects near-misses involving multiple moving objects on a road, and includes a roadside unit installed on or near the road, a sensor provided in the roadside unit, and one or more processors that perform processing related to the near-miss detection based on the detection results of the sensor, wherein the processor detects multiple moving objects on the road around the roadside unit based on the detection results of the sensor, obtains target information related to the moving objects, sets a braking time for any moving object of interest based on the target information and the detection results of the sensor, determines the risk of collision between the moving objects based on the braking time, and determines the near-miss based on the determination results.

[0013] According to this, the braking time for any target moving body is appropriately set based on the target information and the detection results of the sensor. This makes it possible to set appropriate braking times for all related moving bodies, or to set appropriate braking times for any target moving body depending on the situation, thereby improving the accuracy of collision detection and making it possible to appropriately determine near misses.

[0014] In addition, the second invention is configured to obtain at least one of status information regarding the road, status information regarding the moving body, and attribute information regarding the moving body based on the detection results of the sensor, and to set the braking time regarding the moving body of interest based on the target information and at least one of status information regarding the road, attribute information regarding the moving body, and status information regarding the moving body.

[0015] According to this, an appropriate braking time is set for a moving object of interest, taking into consideration the road conditions, the attributes of the moving object, and the state of the moving object.

[0016] In addition, a third invention is configured such that the processor acquires information regarding the road surface condition of the road on which the target moving body is located as status information regarding the road, and sets the braking time based on the information regarding the road surface condition.

[0017] According to this, since the braking time differs depending on the road surface condition (dry, wet, icy, snowy, etc.), the accuracy of near-miss detection is improved by setting the braking time based on the road surface condition. In this case, the road surface condition may be detected based on the detection result of a sensor provided in the roadside device.

[0018] In addition, a fourth invention is configured such that the processor acquires information regarding the type of the target moving body as attribute information regarding the moving body, and sets the braking time based on the information regarding the type of the moving body.

[0019] According to this, since braking times differ depending on the type of moving object (car, motorcycle, bicycle, pedestrian, etc.), setting the braking time based on the type of moving object improves the accuracy of near-miss incident determination. In this case, the type of moving object may be recognized based on the detection results of a sensor installed in the roadside device. Furthermore, information regarding the type of moving object may be provided from the mobile terminal.

[0020] In addition, a fifth invention is configured such that the processor determines that a near miss does not occur if the movement paths of the two target moving bodies do not intersect based on the target information.

[0021] This makes it possible to avoid erroneous determination of near misses. That is, even if a collision between two moving objects is predicted, if the moving paths of the two moving objects do not intersect, for example, if the two moving objects are moving in the same direction, a near miss will not occur, and therefore it will be determined not to be a near miss.

[0022] In addition, a sixth invention is configured such that the processor determines, based on the target information, that a near miss does not occur if a moving body on the priority side passes through the intersection before a moving body on the non-priority side.

[0023] This makes it possible to avoid erroneous determination of near misses. That is, even if a collision between two moving bodies is predicted, if the moving body on the priority side traveling on the priority road passes through the intersection before the moving body on the non-priority side traveling on the non-priority road, a near miss will not occur, and it will be determined that it does not correspond to a near miss.

[0024] In addition, a seventh invention is configured such that the processor measures the braking time of a moving object traveling on a road around the roadside device, stores the measured values ​​of the braking time in a memory unit, obtains statistical information on the braking time on the road around the roadside device based on the stored measured values ​​of the braking time, and sets the braking time for the moving object of interest based on the statistical information and the target information.

[0025] According to this, since the road characteristics that affect braking time differ depending on the location where the roadside unit is installed, by setting the braking time for the moving object that is the subject of collision detection based on statistical information on braking time individually acquired by the roadside unit, the accuracy of collision detection can be improved and near misses can be properly detected.

[0026] In addition, the eighth invention is configured so that the processor detects a specific area that is in an abnormal state, measures the braking time of a moving body passing through the specific area, stores the measured braking time in a memory unit, and when it is predicted that a target moving body will pass through the specific area based on information about the specific area and the target information, sets the braking time for the target moving body based on the measured braking time in the past.

[0027] According to this, when there is a specific area that becomes different from normal due to environmental changes such as weather conditions, the braking time for the moving body that is the subject of collision detection can be set based on the measured braking time of moving bodies that have previously passed through that specific area, thereby improving the accuracy of collision detection and making it possible to properly detect near misses.

[0028] In addition, a ninth invention is a near-miss determination method in which processing related to determining near-misses involving multiple moving objects on a road is executed by one or more processors, and is configured to detect multiple moving objects on the road around the roadside unit based on the detection results of a sensor provided in a roadside unit installed on or near the road, obtain target information related to the moving objects, set a braking time for any moving object of interest based on the target information and the detection results of the sensor, determine the risk of collision between the moving objects based on the braking time, and determine the near-miss based on the determination results.

[0029] According to this, as in the first invention, the braking time for any target moving body is appropriately set based on the target information and the detection results of the sensor, which makes it possible to set appropriate braking times for all related moving bodies or to set appropriate braking times for any target moving body depending on the situation, thereby improving the accuracy of collision detection and making it possible to appropriately determine near misses.

[0030] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0031] First Embodiment FIG. 1 is a diagram showing the overall configuration of a near-miss incident determination system according to a first embodiment.

[0032] The near-miss incident detection system detects near-miss incidents involving moving objects such as vehicles (cars, motorcycles, etc.), pedestrians, and bicycles on a road. The near-miss incident detection system includes a roadside unit 1 and a server 2. The roadside unit 1 and the server 2 can communicate with each other via a network.

[0033] The roadside unit 1 is installed on a road or in its vicinity. The roadside unit 1 includes a sensor 11. The roadside unit 1 uses the sensor 11 to detect targets (moving objects such as vehicles, pedestrians, and bicycles) present on the road around the unit, and generates target information relating to the target's position, speed, traveling direction, size, etc. The sensor 11 may be, for example, a camera, radar, or lidar. The roadside unit 1 may be provided with a plurality of sensors 11 of the same type or different types.

[0034] The roadside device 1 also determines the risk of collision between targets based on the target information collected by the device itself, and predicts a collision between moving objects. The roadside device 1 also provides the collision determination result and the target information to the server 2.

[0035] The server 2 determines whether or not a case in which the roadside unit 1 has determined that there is a high risk of collision between targets in its collision determination (a collision prediction case) corresponds to a near miss, based on the collision determination result and target information acquired from the roadside unit 1. At this time, the collision prediction case becomes a candidate for a near miss, and in the near miss determination performed by the server 2, cases that do not correspond to a near miss are excluded from the collision prediction cases.

[0036] Information related to near-miss incidents is registered in a database and managed in the server 2. An administrator can view information related to near-miss incidents by operating a viewing terminal (not shown) to access the server 2.

[0037] The roadside unit 1 can also communicate with mobile terminals 3 carried by moving objects such as vehicles (cars, motorcycles, etc.), pedestrians, and bicycles using ITS communication, i.e., a communication method defined by ITS (Intelligent Transport Systems). In ITS communication, messages including location information and attribute information of the moving object are transmitted and received between the roadside unit 1 and the mobile terminal 3. In the example shown in FIG. 1 , an in-vehicle terminal serving as the mobile terminal 3 is mounted on the vehicle (car, motorcycle). Pedestrians carry pedestrian terminals serving as the mobile terminal 3. Bicycles also carry bicycle terminals serving as the mobile terminal 3. Note that the following description will mainly focus on cases where the subject of a near-miss incident is a driver of a vehicle (car), but there are also cases where the subject of a near-miss incident is a pedestrian or a bicycle driver.

[0038] Next, the near-miss incident determination performed by the server 2 according to the first embodiment will be described. Fig. 2 is an explanatory diagram showing an overview of the near-miss incident determination.

[0039] The example shown in Figure 2(A) is a case where the movement paths of two targets do not intersect. In this example, the two targets are a vehicle (automobile) and a bicycle, with the vehicle traveling on the roadway and the bicycle traveling on the sidewalk (shoulder strip). In this case, the collision prediction circles of the two targets (vehicle and bicycle) overlap, so the collision detection determines that there is a high risk of collision. Furthermore, if the bicycle suddenly veers into the roadway, the driver of the vehicle may experience a near miss.

[0040] However, because there is a guardrail between the roadway and the sidewalk, the bicycle will not suddenly move onto the roadway and the paths of the vehicle and bicycle will not intersect, so in reality the risk of the two objects colliding is low and the driver of the vehicle will not have a near miss.

[0041] The example shown in FIG. 2(B) is a case where the travel paths of two targets intersect. In this example, the two roads leading to the intersection are divided into priority and non-priority roads. The two targets are a vehicle (automobile) and a bicycle, with the vehicle traveling on the priority road and the bicycle traveling on the non-priority road. In this case, the collision prediction circles of the two targets overlap, and therefore a high risk of collision is determined in the collision detection. Furthermore, if the bicycle traveling on the non-priority road enters the intersection before the vehicle without slowing down or stopping before the intersection, the driver of the vehicle may experience a near miss.

[0042] However, if a vehicle (priority object) traveling on a priority road passes through an intersection before a bicycle (non-priority object) traveling on a non-priority road, the risk of the two objects colliding is actually low, and the driver of the vehicle will not experience a near miss.

[0043] Therefore, in this embodiment, based on the target information, it is determined whether the movement paths of the two target objects intersect, and if the movement paths of the two target objects do not intersect, it is determined that there is no near miss. Also, based on the target information, it is determined whether the priority target object passed through the intersection before the non-priority target object, and if the priority target object passed the intersection before the non-priority target object, it is determined that there is no near miss. In addition to the target information, map information, etc. may also be used as appropriate in the near miss determination.

[0044] There are various types of near-miss incidents, and near-miss incident determination may be performed using appropriate determination criteria depending on the various types of near-miss incidents.

[0045] The collision prediction circle represents the range in which the tip of the target is predicted to arrive after a predetermined time, and is set based on the elapsed time, the target's speed, the target's width, the magnitude of the error in the position information, etc.

[0046] Next, a description will be given of the schematic configuration of the roadside device 1, the server 2, and the mobile terminal 3 according to the first embodiment. Fig. 3 is a block diagram showing the schematic configuration of the roadside device 1, the server 2, and the mobile terminal 3. Fig. 4 is a block diagram showing an overview of the processing performed by the roadside device 1 and the server 2.

[0047] As shown in FIG. 3 , the roadside unit 1 includes a sensor 11 , a wireless communication unit 12 , a network communication unit 13 , a storage unit 14 , and a processor 15 .

[0048] The sensor 11 detects objects present around the roadside unit 1 .

[0049] The wireless communication unit 12 performs ITS communication (road-to-vehicle communication) with the mobile terminal 3 .

[0050] The network communication unit 13 communicates with the server 2 via a network such as the Internet.

[0051] The storage unit 14 stores programs executed by the processor 15 and the like.

[0052] The processor 15 performs various processes by executing the programs stored in the storage unit 14. As shown in Fig. 4, in this embodiment, the processor 15 performs a target detection process, an object target setting process, a TTC calculation process, a road surface condition detection process, a friction coefficient setting process, a braking time setting process, a collision determination process, and the like.

[0053] In the target detection process, the processor 15 detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information. At this time, the position, speed, traveling direction, and size of the target are measured. The target information includes information about the detected target, such as the target ID, time of presence, position, speed, traveling direction, and attributes (size).

[0054] In the target object determination process, the processor 15 determines whether there are two targets (moving bodies) that can be targets for collision determination, and if there are two targets that can be targets for collision determination, the processor 15 sets the two targets as targets for collision determination. At this time, for example, in the vicinity of an intersection, it is determined whether target information has been acquired for each of the two intersecting roads, that is, whether there are targets on each of the two intersecting roads.

[0055] In the TTC calculation process, the processor 15 calculates the TTC (Time to Collision) for two targets, i.e., the remaining time until the two targets collide, based on the target information. At this time, the TTC is calculated based on information such as the positions, speeds, accelerations, and traveling directions of the two targets.

[0056] In the road surface condition detection process, the processor 15 detects the road surface condition (dry, wet, icy, snowy, etc.) of the road on which the target is located based on the detection result of the sensor 11. The road surface condition is detected for each target.

[0057] In the friction coefficient setting process, the processor 15 sets the friction coefficient of the road on which the target object is located based on the road surface condition of the road on which the target object is located. The friction coefficient is set for each target object. For example, the friction coefficient is set to 0.7 to 0.9 when the road surface condition is dry asphalt pavement, and to 0.45 to 0.6 when the road surface condition is wet asphalt pavement.

[0058] In the braking time setting process, the processor 15 sets the braking time of the target, i.e., the time required from the start of braking (brake operation) until the target decelerates and stops, based on the friction coefficient of the road on which the target is located. The braking time is set for each target. The braking time is calculated using the following formula: Braking time = initial speed / (gravitational acceleration × friction coefficient) Here, the initial speed is the speed of the target at the start of braking, and is measured based on the detection result of the sensor 11.

[0059] In the collision determination process, the processor 15 compares the TTC (Time to Collision) of the two targets with the braking time of the priority target to determine whether or not there is a high risk of the two targets colliding with each other. Specifically, if the braking time of the priority target is equal to or longer than the TTC, that is, if the priority target cannot stop by the time the two targets collide with each other, it is determined that there is a high risk of the two targets colliding with each other.

[0060] In addition, in the collision determination process, the processor 15 compares the TTC (time to collision) between two targets with a predetermined threshold (e.g., 0 to 2 seconds) to determine whether there is a high risk of the two targets colliding with each other. Specifically, if the TTC is equal to or less than the threshold, it is determined that there is a high risk of the two targets colliding with each other. In this case, the braking time is not used in the determination, but the TTC may be calculated taking into account the deceleration calculated from the friction coefficient of the road surface as a value related to the braking time. For example, if the road surface is slippery, the friction coefficient becomes small, which reduces the deceleration due to braking, and therefore the TTC becomes shorter.

[0061] The result of the collision determination performed by the roadside device 1 may be transmitted from the roadside device 1 to a mobile terminal 3 (such as an in-vehicle terminal, a pedestrian terminal, or a bicycle terminal) to warn people (drivers and pedestrians) to avoid a collision. The result of the collision determination performed by the roadside device 1 may also be used only for near-miss determination.

[0062] As shown in FIG. 3, the server 2 includes a network communication unit 21, a storage unit 22, and a processor 23.

[0063] The network communication unit 21 communicates with the roadside unit 1 via a network such as the Internet.

[0064] The storage unit 22 stores programs executed by the processor 23. The storage unit 22 also stores target object information collected by the roadside unit 1 and collision determination results acquired by the roadside unit 1. The target object information and collision determination results are registered and managed in a collision prediction case database. The storage unit 22 also stores near-miss determination results acquired by the processor 23. The near-miss determination results are registered and managed in the near-miss case database (see FIG. 4).

[0065] The processor 23 performs various processes by executing programs stored in the storage unit 22. As shown in Fig. 4, in this embodiment, the processor 23 performs near-miss determination processing and the like.

[0066] In the near-miss determination process, the processor 23 determines whether or not a case in which the roadside unit 1 has determined that there is a high risk of collision between moving objects in its collision determination (a collision prediction case) corresponds to a near-miss, based on the collision determination result and target information acquired from the roadside unit 1. If the case corresponds to a near-miss, it is registered as a near-miss case in the near-miss case database. On the other hand, if the case does not correspond to a near-miss, it is excluded from the near-miss cases. Incidents that resulted in accidents are also excluded from the near-miss cases.

[0067] In this embodiment, the state after the time when the two targets collide is observed to determine whether or not it corresponds to a near miss. Specifically, it is determined whether or not the movement paths of the two targets intersect, and if the movement paths of the two targets do not intersect, it is determined that it does not correspond to a near miss. It is also determined whether or not the priority target (a target traveling on a priority road) passes through the intersection before the non-priority target (a target traveling on a non-priority road), and if the priority target passes the intersection before the non-priority target, it is determined that it does not correspond to a near miss.

[0068] The mobile terminal 3 (vehicle terminal, pedestrian terminal, bicycle terminal) includes a positioning unit 31, a wireless communication unit 32, a storage unit 33, and a processor 34.

[0069] The positioning unit 31 uses a satellite positioning system such as GPS to detect the current position of the device itself and acquires position information of the device itself.

[0070] The wireless communication unit 32 performs ITS communication (road-to-vehicle communication) with the roadside unit 1 .

[0071] The storage unit 33 stores programs to be executed by the processor 34. The storage unit 34 also stores information relating to the attributes (type) of the mobile object that owns the device.

[0072] The processor 34 executes the programs stored in the storage unit 33 to perform various processes, such as processes related to the transmission and reception of messages in ITS communication with the roadside unit 1 .

[0073] Next, a description will be given of the processing procedures performed by the roadside device 1 and the server 2 according to the first embodiment. Fig. 5 is a flow diagram showing a first example of the processing procedures performed by the roadside device 1. Fig. 6 is a flow diagram showing a second example of the processing procedures performed by the roadside device 1. Fig. 7 is a flow diagram showing the processing procedures performed by the server 2.

[0074] The first example shown in Fig. 5 is a case where roads are divided into priority and non-priority roads (see Fig. 2(B)). In this case, collision determination is performed by focusing on the braking time of the priority target, i.e., the target traveling on the priority road. Specifically, the braking time of the priority target is set based on the friction coefficient calculated from the road surface condition of the road on which the priority target is located, and in collision determination, the braking time of the priority target is compared with the TTC (Time to Collision).

[0075] In the roadside device 1, first, the processor 15 detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST101).

[0076] Next, the processor 15 determines whether there are two targets that can be the subject of collision determination (target determination process) (ST102).

[0077] Here, if there are two targets that may be subject to collision determination (Yes in ST102), the processor 15 then calculates the TTC (time to collision) for the two targets based on the target information (TTC calculation process) (ST103).

[0078] Next, based on the detection results of the sensor 11, the processor 15 detects the road surface condition (e.g., frozen condition, wet condition) of the road on which the priority target, i.e., the target traveling on the priority road, is located (road surface condition detection process) (ST104).

[0079] Next, the processor 15 sets the friction coefficient of the road on which the priority target is located based on the detected road surface condition (friction coefficient setting process) (ST105).

[0080] Next, the processor 15 sets the braking time for the priority target based on the speed of the priority target and the friction coefficient of the road on which the priority target is located (braking time setting process) (ST106).

[0081] Next, the processor 15 determines whether the braking time of the priority target is equal to or greater than the TTC (time to collision), that is, whether there is a high risk of the two targets colliding with each other (collision determination process) (ST107).

[0082] Here, if the braking time of the priority target is equal to or longer than the TTC, that is, if there is a high risk of the two targets colliding with each other (Yes in ST107), the network communication unit 13 then transmits the collision determination result and target information to the server 2 (ST108).

[0083] The second example shown in Figure 6 is a case where the road is not divided into priority and non-priority sections. In this case, no braking time is set, and collision determination is performed based on the length of the TTC (Time to Collision). Specifically, the TTC (Time to Collision) is calculated based on the deceleration corresponding to the friction coefficient determined from the road surface condition, and in collision determination, the TTC (Time to Collision) is compared with a predetermined threshold. The threshold for the TTC may be set, for example, within a range of 0 to 2 seconds.

[0084] In the roadside device 1, first, the processor 15 detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST201).

[0085] Next, the processor 15 determines whether there are two targets that can be the subject of collision determination (target determination process) (ST202).

[0086] Here, if there are two targets that may be subject to collision determination (Yes in ST202), the processor 15 then detects the road surface conditions of the road on which each of the two target targets is located based on the detection results of the sensor 11 (ST203).

[0087] Next, the processor 15 sets the friction coefficient of the road on which each of the two target objects is located based on the detected road surface condition (friction coefficient setting process) (ST204).

[0088] Next, the processor 15 sets the deceleration of each of the two targets based on the set friction coefficient (deceleration setting process) (ST205).

[0089] Next, the processor 15 calculates the TTC (time to collision) for the two targets based on the set deceleration (TTC calculation process) (ST206).

[0090] Next, the processor 15 determines whether the calculated TTC (time to collision) is equal to or less than a predetermined threshold, that is, whether there is a high risk of the two targets colliding with each other (collision determination process) (ST207).

[0091] Here, if the TTC is equal to or less than the threshold value, i.e., if there is a high risk of the two targets colliding with each other (Yes in ST207), the network communication unit 13 then transmits the collision determination result and target information to the server 2 (ST208).

[0092] 7A, in the server 2, the network communication unit 21 receives the collision determination result and the target object information transmitted from the roadside device 1 (ST301). Next, the processor 23 stores the received information in the storage unit 22 (ST302).

[0093] 7(B), in the server 2, the processor 23 determines whether the movement paths of two targets predicted to collide in the collision determination intersect (ST401) based on the collision determination result and target information stored in the storage unit 22. Here, if the movement paths of the two targets do not intersect (No in ST401), the processor 23 determines that the event does not constitute a near miss (ST404).

[0094] On the other hand, if the movement paths of the two targets intersect (Yes in ST401), processor 23 then determines whether the non-priority target (target traveling on the non-priority road) passed through the intersection before the priority target (target traveling on the priority road) (ST402).Here, if the non-priority target did not pass through the intersection before the priority target (No in ST402), processor 23 determines that this does not constitute a near miss (ST404).

[0095] On the other hand, if the non-priority target passes the intersection before the priority target (Yes in ST402), processor 23 determines that this corresponds to a near miss (ST403). Next, processor 23 registers the target information as a near miss case in the near miss case database. Note that if it is determined that the case is not a near miss, the information of the non-near miss case may be registered in another database, or the information of the non-near miss case may be registered in the same database as the near miss case so as to be distinguishable from the near miss case.

[0096] In this embodiment, the roadside device 1 performs collision detection (collision prediction) in real time, and the server 2 performs near-miss detection afterward by batch processing, but the near-miss detection may be performed in real time by the roadside device 1 following the collision detection. Also, the server 2 may perform collision detection and near-miss detection afterward by batch processing.

[0097] In this embodiment, collision detection is performed by comparing the braking time with the TTC (Time to Collision) or by comparing the TTC with a threshold value, but in addition to such time detection, collision detection can also be performed by comparing positions, for example, by comparing a position where the vehicle can be stopped by braking with a predicted collision position. Collision detection may also be performed based on information about a positional range, such as a collision prediction circle.

[0098] In this embodiment, the processor 15 of the roadside unit 1 acquires the road surface condition of the road on which the target object is located as road condition information, and sets the braking time based on the road surface condition, but the road condition information is not limited to the road surface condition.

[0099] For example, the processor 15 of the roadside unit 1 may acquire the gradient of the road on which the target object is located as road status information and set the braking time based on the gradient. In this case, the processor 15 of the roadside unit 1 can acquire the gradient of the road based on map information. The processor 15 of the roadside unit 1 can also measure the gradient (inclination) of the road based on the detection result of the sensor 11 provided in the roadside unit 1.

[0100] The processor 15 of the roadside unit 1 may also acquire information about obstacles (frozen areas, snow-covered areas, etc.) on the road where the target object is located as road condition information, and set the braking time based on the information about the obstacles. In this case, the processor 15 of the roadside unit 1 can detect obstacles (frozen areas, snow-covered areas, etc.) on the road based on the detection results of the sensor 11 provided in the roadside unit 1.

[0101] The processor 15 of the roadside unit 1 may also acquire information about weather conditions (fog, blizzard, etc.) around the roadside unit 1 as road condition information, and set the braking time based on the information about the weather conditions. In this case, the processor 15 of the roadside unit 1 can detect the weather conditions around the roadside unit 1 based on the detection results of the sensor 11 provided in the roadside unit 1. The processor 15 of the roadside unit 1 may also acquire information about weather conditions around the roadside unit 1 from a weather information server via a network.

[0102] The processor 15 of the roadside device 1 may set the braking time based on one type of road-related condition information, but may also set the braking time by combining multiple types of road-related condition information.

[0103] Second Embodiment Next, a second embodiment will be described. It should be noted that the points not specifically mentioned here are the same as those in the previous embodiment. Fig. 8 is a block diagram showing an outline of the processing performed by the roadside device 1 and the server 2 according to the second embodiment.

[0104] The braking time differs depending on the type of target (moving object) (vehicle (car, motorcycle, etc.), bicycle, pedestrian, etc.). For example, a pedestrian can stop quickly, but the braking time of a vehicle (car) becomes longer depending on its speed. Therefore, in this embodiment, the processor 15 of the roadside unit 1 recognizes the type of target based on the detection result of the sensor 11, and sets the braking time according to the type of target to perform collision determination. Note that the type of moving object is classified according to the shape and size of the moving object; for example, cars may be further classified into large vehicles and small vehicles.

[0105] In this embodiment, the processor 15 of the roadside device 1 performs target detection processing, object target setting processing, target recognition processing, TTC calculation processing, braking time setting processing, collision determination processing, etc. The target detection processing, object target setting processing, TTC calculation processing, and collision determination processing are the same as those in the first embodiment.

[0106] In the target recognition process, the processor 15 recognizes the type (car, motorcycle, bicycle, pedestrian, etc.) of the target (moving object) detected in the target detection process based on the detection result of the sensor 11. Specifically, for example, the type of target is recognized by image recognition processing of an image captured by a camera serving as the sensor 11. The target recognition process may be performed using an object recognition engine created by machine learning such as deep learning. Alternatively, the target recognition process may be performed using an image recognition technique such as pattern matching.

[0107] In the braking time setting process, the processor 15 sets the braking time for the priority target based on the speed and type of the priority target. Information (e.g., a table) relating to the correspondence relationship between the speed and type of the target and the braking time is stored in advance in the memory unit 14, and the processor 15 can determine the braking time from the speed and type of the target by referring to the information relating to the correspondence relationship.

[0108] Next, a description will be given of the processing procedure performed by the roadside device 1 according to the second embodiment. Fig. 9 is a flow diagram showing the processing procedure performed by the roadside device 1. Note that the processing performed by the server 2 is the same as that in the first embodiment (see Fig. 7).

[0109] In the roadside device 1, first, the processor 15 detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST101).

[0110] Next, the processor 15 recognizes the type (car, motorcycle, bicycle, pedestrian, etc.) of the target (moving object) detected in the target detection process based on the detection result of the sensor 11 (target recognition process) (ST111).

[0111] Next, the processor 15 determines whether there are two targets that can be the subject of collision determination (target determination process) (ST102).

[0112] Here, if there are two targets that may be subject to collision determination (Yes in ST102), the processor 15 then calculates the TTC (time to collision) for the two targets based on the target information (TTC calculation process) (ST103).

[0113] Next, the processor 15 sets the braking time of the priority target of the two target objects, i.e., the target traveling on the priority road, based on the speed and the type of the priority target (braking time setting process) (ST112).

[0114] Next, the processor 15 determines whether the braking time of the priority target is equal to or greater than the TTC (time to collision), that is, whether there is a high risk of the two targets colliding with each other (collision determination process) (ST107).

[0115] Here, if the braking time of the priority target is equal to or longer than the TTC, that is, if there is a high risk of the two targets colliding with each other (Yes in ST107), the network communication unit 13 then transmits the collision determination result and target information to the server 2 (ST108).

[0116] 9 is a case where roads are divided into priority and non-priority (see FIG. 2B), and processing is performed in the same procedure as in the first example in the first embodiment (see FIG. 5). However, there are also cases where roads are not divided into priority and non-priority, and in this case processing is performed in the same procedure as in the second example in the first embodiment (see FIG. 6). This also applies to the modified examples and other embodiments described below.

[0117] In this embodiment, the processor 15 of the roadside unit 1 acquires the type of the target as attribute information about the target (moving object) and sets the braking time based on the type of the target, but the braking time may also be set based on status information about the target.

[0118] For example, the processor 15 of the roadside device 1 may acquire the state of the tires mounted on the vehicle as the target as the state information related to the target, and set the braking time based on the state of the tires (e.g., whether chains are mounted, etc.). In this case, the processor 15 of the roadside device 1 may detect the state of the tires based on the detection result of the sensor 11 provided in the roadside device 1 (e.g., an image captured by a camera).

[0119] The processor 15 of the roadside unit 1 may set the braking time based on only one of the attribute information and the status information related to the target object, or may set the braking time by combining multiple types of attribute information and status information related to the target object. Furthermore, the processor 15 of the roadside unit 1 may set the braking time by combining status information related to the road (e.g., road surface conditions) as in the first embodiment with at least one of the attribute information and the status information related to the target object.

[0120] (Modification of the Second Embodiment) Next, a modification of the second embodiment will be described. Note that points not specifically mentioned here are the same as those in the above-described embodiment. Fig. 10 is a block diagram showing an outline of the processing performed by the roadside device 1, the server 2, and the mobile terminal 3 according to the modification of the second embodiment.

[0121] In the second embodiment, the roadside device 1 recognizes the type (automobile, motorcycle, bicycle, pedestrian, etc.) of the target (moving object) detected in the target detection process based on the detection result of the sensor 11. On the other hand, in this modification, information on the type of the target is provided from the mobile terminal 3 to the roadside device 1. Specifically, the information on the type of the target is added to a message of ITS communication (road-to-vehicle communication) as target information (moving object information), and is transmitted from the mobile terminal 3 to the roadside device 1. Note that the mobile terminal 3 stores information on the type of the moving object that owns the device in advance in the storage unit 33.

[0122] In this embodiment, the processor 15 of the roadside device 1 performs target detection processing, object target setting processing, target recognition processing, TTC calculation processing, braking time setting processing, collision determination processing, etc. The target detection processing, object target setting processing, TTC calculation processing, braking time setting processing, and collision determination processing are the same as those in the above-described embodiment.

[0123] In the target recognition process, the processor 15 recognizes the type (car, motorcycle, bicycle, pedestrian, etc.) of the target (moving object) detected in the target detection process, based on the target information (moving object information) provided by the mobile terminal 3. At this time, the processor 15 compares the position of the target detected in the target detection process with the position information of the moving object included in the ITS communication message transmitted from the mobile terminal 3, and associates the detected target with the moving object that transmitted the message, thereby obtaining the type of the detected target.

[0124] Next, a description will be given of the processing procedure performed by the roadside device 1 and the mobile terminal 3 according to the modified example of the second embodiment. Fig. 11 is a flow diagram showing the processing procedure performed by the roadside device 1 and the mobile terminal 3. The processing performed by the server 2 is the same as that in the first embodiment (see Fig. 7).

[0125] 11A, in the mobile terminal 3 (vehicle terminal, pedestrian terminal, bicycle terminal), the wireless communication unit 32 periodically transmits target information to the roadside unit 1 (ST102). The target information includes information about the type of the mobile object, and is added to an ITS communication message and transmitted to the roadside unit 1.

[0126] As shown in FIG. 11B, in the roadside device 1, first, the processor 15 detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST101).

[0127] In addition, in the roadside device 1, the wireless communication unit 12 receives the target information transmitted from the mobile terminal 3 (ST121).

[0128] Next, in the roadside device 1, the processor 15 recognizes the type (car, motorcycle, bicycle, pedestrian, etc.) of the target (moving object) detected in the target detection process based on the target information received from the mobile terminal 3 (target recognition process) (ST122).

[0129] The subsequent processing is the same as in the second embodiment (see FIG. 9).

[0130] Third Embodiment Next, a third embodiment will be described. It should be noted that the points not specifically mentioned here are the same as those in the above-described embodiments. Fig. 12 is a block diagram showing an outline of the processing performed by the roadside device 1 and the server 2 according to the third embodiment.

[0131] The characteristics of the road that affect the braking time differ depending on the location where the roadside unit 1 is installed. Therefore, in this embodiment, the braking time of the target when the target passed through the road around the roadside unit 1 in the past is measured, and the measured values ​​of the braking time are statistically processed to generate statistical information that reflects the characteristics of the road around the roadside unit 1. The statistical information is then used to set the braking time for the target that is the subject of collision judgment.

[0132] In this embodiment, the processor 15 of the roadside device 1 performs target detection processing, road surface condition detection processing, braking time measurement processing, statistical processing, target setting processing, TTC calculation processing, braking time setting processing, collision determination processing, etc. The target detection processing, road surface condition detection processing, target setting processing, TTC calculation processing, and collision determination processing are the same as those in the above-described embodiment.

[0133] In the braking time measurement process, the processor 15 measures the braking time of the target, i.e., the time from the start of braking (brake operation) to deceleration and stop, based on the target information acquired in the target detection process. The braking time is measured for each target detected in the target detection process. The braking time acquired in the braking time measurement process is stored in the memory unit 14 in association with the target trajectory, target speed, and road surface condition of the road on which the target is located at the time the braking time was measured.

[0134] In the statistical processing, the processor 15 generates statistical information that reflects the characteristics of the road around the roadside unit 1 based on the information stored in the memory unit 14, i.e., the measured braking time of the target when it passed through the road around the roadside unit 1 in the past, the trajectory of the target when the braking time was measured, the speed of the target, and the road surface condition of the road on which the target is located.

[0135] In the braking time setting process, the processor 15 sets the braking time of the target object based on the statistical information, the target object information, and the information on the road surface condition. At this time, by using the statistical information, the braking time of the target object can be calculated from the current trajectory and speed of the target object and the current road surface condition of the road on which the target object is located.

[0136] Next, a description will be given of the processing procedure performed by the roadside device 1 according to the third embodiment. Fig. 13 is a flow diagram showing the processing procedure performed by the roadside device 1. Note that the processing performed by the server 2 is the same as that in the first embodiment (see Fig. 7).

[0137] 13A, the information collection process will be described. In the roadside device 1, the processor 15 first detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST131).

[0138] Next, the processor 15 detects the road surface condition (for example, frozen condition, wet condition) of the road on which the target is located based on the detection result of the sensor 11 (road surface condition detection process) (ST132).

[0139] Furthermore, the processor 15 measures the braking time of the target based on the target information (braking time measurement process) (ST133).

[0140] Next, the processor 15 stores the measured braking time, the trajectory of the target at the time the braking time was measured, the speed of the target, and the road surface condition of the road on which the target is located in the memory unit 14 (collected information storage process) (ST134).

[0141] 13B , the processor 15 of the roadside unit 1 acquires statistical information about braking times on roads around the roadside unit 1 based on the information stored in the memory unit 14, i.e., the measured values ​​of braking times, the trajectory of the target object at the time the braking times were measured, the speed of the target object, and the road surface conditions of the road on which the target object is located (statistical processing) (ST141).

[0142] 13C , the roadside device 1 first detects targets (moving objects) on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST101).

[0143] Next, the processor 15 determines whether there are two targets that can be the subject of collision determination (target determination process) (ST102).

[0144] Here, if there are two targets that may be subject to collision determination (Yes in ST102), the processor 15 then calculates the TTC (time to collision) for the two targets based on the target information (TTC calculation process) (ST103).

[0145] Next, based on the detection results of the sensor 11, the processor 15 detects the road surface condition (e.g., frozen condition, wet condition) of the road on which the priority target, i.e., the target traveling on the priority road, is located (road surface condition detection process) (ST104).

[0146] Next, the processor 15 sets the braking time of the priority target based on the road surface condition of the road on which the priority target is located, the statistical information, and the target information (braking time setting process) (ST151). At this time, the braking time of the priority target is calculated from the current trajectory and speed of the priority target and the road surface condition of the road on which the priority target is located, with reference to the statistical information.

[0147] Next, the processor 15 determines whether the braking time of the priority target is equal to or greater than the TTC (time to collision), that is, whether there is a high risk of the two targets colliding with each other (collision determination process) (ST107).

[0148] Here, if the braking time of the priority target is equal to or longer than the TTC, that is, if there is a high risk of the two targets colliding with each other (Yes in ST107), the network communication unit 13 then transmits the collision determination result and target information to the server 2 (ST108).

[0149] In this embodiment, once a certain amount of statistical information has been accumulated by the information collection process shown in Fig. 13(A) and the analysis process shown in Fig. 13(B), the information collection process shown in Fig. 13(A), the analysis process shown in Fig. 13(B), and the operation process shown in Fig. 13(C) may be executed in parallel. This allows the statistical information to be updated as needed during operation, improving the accuracy of collision determination.

[0150] (Fourth embodiment) Next, a fourth embodiment will be described. Note that points not specifically mentioned here are the same as those in the above-described embodiments. Fig. 14 is an explanatory diagram showing an overview of the specific area detection process performed by the server 2 according to the fourth embodiment.

[0151] On roads around the roadside unit 1, environmental changes such as weather conditions can cause areas where mobile objects cannot pass through as usual, such as slippery areas due to ice or snow. Also, areas where passage is hindered due to damaged pavement can appear. Such areas become obstacles for mobile objects to pass through, causing the mobile objects to behave differently from usual and braking times to differ from usual.

[0152] In the example shown in Figure 14, slippery areas due to ice or snow have appeared on the roadway. In this case, pedestrians walk carefully to avoid the slippery areas when crossing the roadway. Vehicles also try to slow down to pass through the slippery areas. When an area appears in which moving objects cannot pass through normally, the moving objects behave differently from usual, and braking times also differ from usual.

[0153] Therefore, in this embodiment, specific areas that become different from normal due to environmental changes such as weather conditions are detected based on the trajectory and speed changes of the moving body, and measurement values ​​such as the braking time of the moving body when it passes through that specific area are accumulated, and these measurement values ​​are used to set the braking time for the target that is the subject of collision judgment.

[0154] Next, an outline of the processing performed by the roadside device 1 and the server 2 according to the fourth embodiment will be described. FIG. 15 is a block diagram showing an outline of the processing performed by the roadside device 1 and the server 2.

[0155] In this embodiment, the processor 15 of the roadside device 1 performs target detection processing, specific area detection processing, road surface condition detection processing, braking time measurement processing, target setting processing, TTC calculation processing, braking time setting processing, collision determination processing, etc. The target detection processing, target setting processing, TTC calculation processing, and collision determination processing are the same as those in the above-described embodiment.

[0156] In the specific area detection process, the processor 15 detects a specific area (e.g., an area that is slippery due to ice or snow) that is in an abnormal state due to environmental changes such as weather conditions, based on the target information and the detection results of the sensor 11 (e.g., a camera image). At this time, for example, an area where the target behaves abnormally is detected as the specific area. Specifically, an area where the target takes a detour is detected as the specific area based on the target's trajectory. Furthermore, an area where an abnormal speed change occurs is detected as the specific area based on the target's speed. Furthermore, when an obstacle (frozen area, snow-covered area) that appears on the road is recognized by image recognition technology, the obstacle is detected as the specific area.

[0157] In the braking time measurement process, the processor 15 measures the braking time of the target, i.e., the time from the start of braking (brake operation) to deceleration and stop, based on the target information acquired in the target detection process. The braking time is measured for each target detected in the target detection process. The braking time acquired in the braking time measurement process is stored in the memory unit 14 in association with the target speed at the time the braking time was measured and the road surface condition of the road on which the target is located.

[0158] In the braking time setting process, the processor 15 sets the braking time of the priority target that is the subject of collision judgment based on the information stored in the memory unit 14, i.e., information such as measured values ​​of braking times for targets that have passed through the specific area in the past, and the target information. At this time, cases similar to the current situation in terms of the speed of the priority target and the road surface condition of the road on which the priority target is located are searched for, and the braking time of the priority target is calculated based on the braking time of the similar case.

[0159] Next, a process performed by the roadside device 1 according to the fourth embodiment will be described. Fig. 16 is a flow chart showing the process performed by the roadside device 1. The process performed by the server 2 is the same as that in the first embodiment (see Fig. 7).

[0160] 16A, the information collection process will be described. In the roadside device 1, the processor 15 first detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST131).

[0161] Next, the processor 15 determines, based on the target information, whether there is a specific area (for example, an area that is slippery due to ice or snow) where the target behaves differently from normal (specific area detection process) (ST161).

[0162] Here, if there is a specific area where the target is behaving differently from normal (Yes in ST161), the processor 15 then detects the road surface condition (e.g., icy condition, snowy condition) of the road where the target is located based on the detection results of the sensor 11 (road surface condition detection process) (ST132).

[0163] Furthermore, the processor 15 measures the braking time when the target passes through the specific area based on the target information (braking time measurement process) (ST133).

[0164] Next, the processor 15 stores the measured braking time, the target's speed when the braking time was measured, and the road surface condition of the road on which the target is located in the memory unit 14 (collected information storage process) (ST134).

[0165] 16(B) will be described. In the roadside device 1, first, the processor 15 detects targets (moving objects) present on the road around the device based on the detection results of the sensor 11, and generates target information (target detection process) (ST101).

[0166] Next, the processor 15 determines whether there are two targets that can be the subject of collision determination (target setting process) (ST102).

[0167] Here, if there are two targets that may be subject to collision determination (Yes in ST102), the processor 15 then calculates the TTC (time to collision) for the two targets based on the target information (TTC calculation process) (ST103).

[0168] Next, based on the detection results of the sensor 11, the processor 15 detects the road surface condition (e.g., icy condition, snowy condition) around the priority target, i.e., the target traveling on the priority road, of the two target objects (road surface condition detection process) (ST104).

[0169] Next, the processor 15 determines whether or not the priority target is predicted to pass through the specific area based on the trajectory and speed of the priority target (ST171).

[0170] If it is predicted that the priority target will pass through the specific area (Yes in ST171), the processor 15 then executes a braking time setting process for the specific area (ST172). At this time, the processor 15 sets the braking time for the priority target based on information such as the measured value of the braking time stored in the memory unit 14 and the target information. On the other hand, if it is not predicted that the priority target will pass through the specific area (No in ST171), the processor 15 then executes a braking time setting process for the non-specific area (ST173). At this time, the processor 15 sets the braking time for the priority target using the same procedure as in the first embodiment (ST106 in FIG. 5).

[0171] Next, the processor 15 determines whether the braking time of the priority target is equal to or greater than the TTC (time to collision), that is, whether there is a high risk of the two targets colliding with each other (collision determination process) (ST107).

[0172] Here, if the braking time of the priority target is equal to or longer than the TTC, that is, if there is a high risk of the two targets colliding with each other (Yes in ST107), the network communication unit 13 then transmits the collision determination result and target information to the server 2 (ST108).

[0173] As described above, the embodiments have been described as examples of the technology disclosed in this application. However, the technology in this disclosure is not limited to these, and can be applied to embodiments in which modifications, substitutions, additions, omissions, etc. are made. Furthermore, it is also possible to combine the components described in the above embodiments to create new embodiments.

[0174] The near miss detection system and near miss detection method disclosed herein have the effect of improving the accuracy of collision detection and appropriately detecting near misses by setting appropriate braking times for all moving bodies involved in collision detection that takes braking times into account, or by setting appropriate braking times for any moving body of interest depending on the situation, and are useful as a near miss detection system that detects near misses involving multiple moving bodies on a road, and a near miss detection method in which processing related to near miss detection is executed by a processor.

[0175] 1: Roadside unit 2: Server 3: Mobile terminal 11: Sensor 14: Storage unit 15: Processor 22: Storage unit 23: Processor

Claims

1. A near-miss detection system that detects near-misses involving multiple moving objects on a road, comprising: a roadside unit installed on or near a road; a sensor provided in the roadside unit; and one or more processors that execute processing related to the near-miss detection based on the detection results of the sensor, wherein the processor: detects multiple moving objects on the road around the roadside unit based on the detection results of the sensor and acquires target information related to the moving objects; sets a braking time for any moving object of interest based on the target information and the detection results of the sensor; determines the risk of collision between the moving objects based on the braking time; and determines the near-miss based on the determination results.

2. The near-miss judgment system described in claim 1, characterized in that the processor acquires at least one of status information regarding the road, status information regarding the moving body, and attribute information regarding the moving body based on the detection results of the sensor, and sets the braking time for the moving body of interest based on the target information and at least one of the status information regarding the road, attribute information regarding the moving body, and status information regarding the moving body.

3. The near-miss judgment system described in claim 2, characterized in that the processor acquires information regarding the road surface condition of the road on which the target mobile object is located as the road condition information, and sets the braking time based on the information regarding the road surface condition.

4. The near-miss judgment system described in claim 2, characterized in that the processor acquires information regarding the type of target mobile body as attribute information regarding the mobile body, and sets the braking time based on the information regarding the type of mobile body.

5. The near-miss determination system described in claim 1, characterized in that the processor determines that a near-miss does not occur if the movement paths of the two target moving objects do not intersect based on the target information.

6. The near-miss determination system described in claim 1, characterized in that the processor determines that a near-miss does not occur if a moving object on the priority side passes through the intersection before a moving object on the non-priority side based on the target information.

7. The near-miss judgment system described in claim 1, characterized in that the processor measures the braking time of mobile objects traveling on roads around the roadside unit, stores the measured values ​​of the braking time in a memory unit, obtains statistical information regarding the braking time on roads around the roadside unit based on the stored measured values ​​of the braking time, and sets the braking time for the target mobile object based on the statistical information and the target information.

8. The near-miss judgment system described in claim 1, characterized in that the processor detects a specific area that is in an abnormal state, measures the braking time of a moving object passing through the specific area, and stores the measured braking time in a memory unit, and when it is predicted that a target moving object will pass through the specific area based on information about the specific area and the target information, sets the braking time for the target moving object based on the measured braking time in the past.

9. A near-miss determination method in which processing related to determining near-misses involving multiple moving objects on a road is executed by one or more processors, the near-miss determination method comprising the steps of: detecting multiple moving objects on the road around a roadside unit based on the detection results of a sensor provided in the roadside unit installed on or near the road, acquiring target information related to the moving objects; setting a braking time for any moving object of interest based on the target information and the detection results of the sensor; determining the risk of collision between the moving objects based on the braking time; and determining the near-miss based on the determination results.

Images