Obstacle sensing device

Abstract

Provided is an obstacle sensing device capable of sensing an obstacle on a road with high accuracy. An obstacle sensing device is provided with: an acquisition unit that acquires, from each of a plurality of vehicles traveling in a predetermined direction on a road, behavior data indicating the behavior of the vehicle, position information indicating the position of the vehicle, and vehicle speed information indicating the speed of the vehicle; a recognition unit that recognizes, on the basis of the behavior data, an avoidance behavior of the vehicle to avoid an object on the road; a detection unit that detects, on the basis of the position information, a position on the road at which the frequency at which the avoidance behavior is recognized by the recognition unit is equal to or greater than a threshold value; a determination unit that determines, on the basis of the vehicle speed information, whether or not the distribution of the speed of the vehicle in the predetermined direction side of the position detected by the detection unit exhibits bimodality; and a sensing unit that does not sense the occurrence of an obstacle at the position detected by the detection unit when the determination unit determines that the distribution exhibits bimodality, and that senses the occurrence of an obstacle at the position detected by the detection unit when the determination unit determines that the distribution does not exhibit bimodality.

Description

Technical Field

[0001] This invention relates to an obstacle sensing (detection) device. Background Technology

[0002] Regarding technologies for sensing obstacles on the road, for example, Patent Document 1 describes the following: based on data representing the behavior (actions) of multiple vehicles in motion, it is determined whether a vehicle has attempted to avoid an obstacle on the road, thereby sensing the obstacle.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2024-88500 Summary of the Invention

[0006] The problem that the invention aims to solve

[0007] Based on the above method, for example, it may falsely sense obstacles because it will detect behaviors such as lane changes at highway exits and service area entrances to avoid congestion.

[0008] Therefore, the present invention was made in view of the above-mentioned problems, and the object is to provide an obstacle sensing device capable of sensing obstacles on a road with high accuracy.

[0009] Technical solutions for solving the problem

[0010] The obstacle sensing device of the present invention comprises: an acquisition unit that acquires behavior data representing the behavior of a plurality of vehicles traveling in a predetermined direction on a road, position information representing the position of the vehicles, and vehicle speed information representing the speed of the vehicles; an identification unit that identifies, based on the behavior data, the avoidance behavior of the vehicles avoiding objects on the road; a detection unit that detects, based on the position information, positions on the road where the frequency of the avoidance behavior identified by the identification unit is above a threshold; a determination unit that determines, based on the vehicle speed information, whether the speed distribution of the vehicles on the predetermined direction side of the position detected by the detection unit exhibits bimodality; and a sensing unit that, if the determination unit determines that the distribution exhibits bimodality, does not sense an obstacle at the position detected by the detection unit, and if the determination unit determines that the distribution does not exhibit bimodality, senses an obstacle at the position detected by the detection unit.

[0011] Invention Effects

[0012] According to the present invention, obstacles on the road can be sensed with high precision. Attached Figure Description

[0013] Figure 1This is a diagram illustrating an example of an obstacle sensing system.

[0014] Figure 2 (a) is a diagram illustrating an example of vehicle behavior and speed distribution information when an obstacle occurs on the road. Figure 2 (b) is a diagram showing an example of vehicle behavior and speed distribution information when there are no obstacles on the road.

[0015] Figure 3 This is a flowchart illustrating an example of the actions of an obstacle sensing server.

[0016] Explanation of reference numerals in the attached figures

[0017] 1 Obstacle sensing server (obstacle sensing device); 10 CPU; 101 Acquisition unit; 102 Recognition unit; 103 Detection unit; 104 Judgment unit; 105 Sensing unit; 133 Speed ​​distribution information. Detailed Implementation

[0018] (Composition of an obstacle sensing system)

[0019] Figure 1 This is a configuration diagram illustrating an example of an obstacle sensing system S. The obstacle sensing system S includes an obstacle sensing server 1 and multiple vehicle-mounted devices 2.

[0020] The vehicle-mounted device 2 is installed in vehicles such as hybrid vehicles and electric vehicles. The vehicle-mounted device 2 includes a data communication module (DCM) 20, a yaw sensor 21, a vehicle speed sensor 22, and a GPS (Global Positioning System) 23.

[0021] Yaw sensor 21 detects the angular acceleration, or yaw rate, of the vehicle's yaw direction. Vehicle speed sensor 22 detects the vehicle's speed. GPS 23 detects the vehicle's latitude and longitude.

[0022] The DCM5 connects to a communication network (not shown) such as the Internet via a wireless communication unit. The DCM5 sends the detection values ​​from the yaw sensor 21, the vehicle speed sensor 22, and the GPS 23 to the obstacle sensing server 1.

[0023] Obstacle sensing server 1 is an example of an obstacle sensing device. Obstacle sensing server 1 has a CPU (Central Processing Unit) 10, ROM (Read Only Memory) 11, RAM (Random Access Memory) 12, SSD (Solid State Drive) 13 and a communication port 14 interconnected via bus 19.

[0024] ROM11 stores the program that drives CPU10. RAM12 functions as the working memory of CPU10. Communication port 14 is connected to a communication network.

[0025] When the CPU 10 reads a program from the ROM 11, it forms a control unit 100, an acquisition unit 101, an identification unit 102, a detection unit 103, a determination unit 104, and a sensing unit 105 as software functions. Additionally, the SSD 13 stores behavior data 130, position information 131, vehicle speed information 132, and speed distribution information 133. The control unit 100 instructs the acquisition unit 101, identification unit 102, detection unit 103, determination unit 104, and sensing unit 105 to operate according to the program sequence.

[0026] The acquisition unit 101 acquires behavior data 130, position information 131, and vehicle speed information 132 from multiple vehicles traveling on the road in a predetermined direction of travel. The behavior data 130, position information 131, and vehicle speed information 132 are acquired from the yaw sensor 21, GPS 23, and vehicle speed sensor 22, respectively.

[0027] The recognition unit 102 identifies avoidance behaviors of the vehicle when avoiding objects on the road based on the behavior data 130. For example, the recognition unit 102 identifies road changes and lane changes as avoidance behaviors based on the time changes of each acceleration of yaw.

[0028] The detection unit 103 detects locations on the road where the frequency of avoidance behavior identified by the recognition unit 102 is above a threshold, based on the location information 131. For example, the detection unit 103 stores the number of avoidance behaviors detected within a predetermined time period in association with the location information 131.

[0029] The determination unit 104 determines, based on the vehicle speed information 132, whether the speed distribution of vehicles on the direction of travel on the road at the location detected by the detection unit 103 exhibits a bimodal distribution. For example, based on the vehicle speed information 132 and the location information 131, the determination unit 104 generates speed distribution information 133, representing the speed distribution of vehicles passing through within a predetermined time for each section of the road, and determines whether a bimodal distribution exists based on the speed distribution information 133.

[0030] If the determination unit 104 determines that the velocity distribution exhibits a bimodal pattern, the sensing unit 105 does not detect an obstacle at the location (detection location) detected by the detection unit 103. If the determination unit 104 determines that the velocity distribution does not exhibit a bimodal pattern, the sensing unit 105 detects an obstacle at the detection location.

[0031] (Example of obstacle sensing)

[0032] Figure 2 Figure (a) is an example of the behavior and speed distribution information 133 of vehicles V and Vx when an obstacle occurs on road R. Road R has two lanes, L1 and L2, and multiple vehicles V and Vx travel in the direction of travel D. Each vehicle V and Vx is equipped with an onboard device 2.

[0033] There is an obstacle X, such as a disabled vehicle or sand, in lane L1. Therefore, vehicle Vx in lane L1 changes lanes to lane L2 to avoid obstacle X (see attached diagram, label K). Subsequently, vehicles V following vehicle Vx do the same. As a result, congestion occurs in lanes L1 and L2, causing a decrease in the speed of vehicles V and Vx ahead of obstacle X.

[0034] The identification unit 102 identifies the avoidance behavior of vehicle Vx. The detection unit 103 calculates the frequency as the number of times the avoidance behavior is identified at position Pa on road R within a predetermined time. Based on the position information 131, the detection unit 103 detects positions Pa with a frequency above a threshold.

[0035] The determination unit 104 generates speed distribution information 133 for vehicle V at position Pb on the direction of travel, passing through position Pa detected by the detection unit 103, based on vehicle speed information 132 and position information 131. In this example, since congestion caused by obstacle X is occurring, the frequency of low speeds is high in both lanes L1 and L2, which is represented as a peak P0 in the speed distribution information 133. Therefore, the determination unit 104 determines that the speed distribution information 133 does not have a bimodal distribution, and the sensing unit 105 detects an obstacle on the road R.

[0036] Figure 2 Figure (b) is an example of the behavior and speed distribution information of vehicles V and Vx on road R when there are no obstacles. Lane L1 is, for example, a road leading to a highway exit or service area entrance, and is therefore congested. Vehicle Vx changes lanes to another lane L2 to avoid the congestion (see attached diagram label K).

[0037] The detection unit 103 detects positions Pc with frequencies above a threshold based on position information 131. The determination unit 104 generates speed distribution information 133 of the vehicle V, which passes through the position Pd on the direction of travel of the position Pc detected by the detection unit 103, based on vehicle speed information 132 and position information 131. In this example, because lane L1 is congested, the frequency of low speeds is high, which is represented by peak P1 in speed distribution information 133. On the other hand, because lane L2 is not congested, the frequency of high speeds is high, which is represented by another peak P2 in speed distribution information 133. Therefore, the determination unit 104 determines that speed distribution information 133 has bimodality, and the sensing unit 105 does not detect any obstacles on the road R. In addition, the determination unit 104 can, for example, detect the boundary speed P3 between high and low speeds, and determine bimodality based on the frequency difference between the boundary speed P3 and peaks P1 and P2, and the speed difference between peaks P1 and P2.

[0038] (Operations of the obstacle detection server)

[0039] Figure 3 This is a flowchart illustrating an example of the operation of obstacle sensing server 1. This operation is performed, for example, at regular intervals and includes the following steps St1 to St8.

[0040] First, the acquisition unit 101 acquires behavior data 130, location information 131, and vehicle speed information 132 from the vehicle-mounted device 2 (St1). Next, the recognition unit 102 determines whether the vehicle's avoidance behavior has been recognized (St2). If it has not been recognized (St2: No), this operation ends.

[0041] If an avoidance behavior is detected (St2: Yes), the detection unit 103 detects the location where the avoidance behavior was detected (St3). The detection unit 103 compares the detection frequency at this location with a predetermined threshold (St4). If the frequency is less than or equal to the threshold (St4: No), this operation ends.

[0042] If the frequency is greater than a threshold (St4: Yes), speed distribution information 133 of the vehicle ahead of the detected position is generated (St5). Next, the determination unit 104 determines whether the speed distribution information 133 has bimodal characteristics (St6). If there is no bimodal characteristic (St6: Yes), the sensing unit 105 detects an obstacle (St7). If there is bimodal characteristics (St6: No), the sensing unit 105 does not detect an obstacle (St8). In this way, the obstacle detection server 1 operates.

[0043] As described above, obstacle sensing server 1 detects obstacles based on the presence or absence of a bimodal distribution in the velocity distribution ahead of locations where avoidance behavior frequently occurs. Therefore, it can distinguish between congestion caused by obstacle X and congestion at highway exits or service area entrances. Thus, obstacle sensing server 1 can detect obstacles on the road with high accuracy.

Claims

1. An obstacle sensing device, comprising: The acquisition unit acquires behavioral data representing the behavior of multiple vehicles traveling in a predetermined direction on the road, position information representing the position of the vehicles, and speed information representing the speed of the vehicles, respectively. The identification unit identifies the vehicle's avoidance behavior when avoiding objects on the road based on the behavioral data. The detection unit detects locations on the road where the frequency of the avoidance behavior identified by the recognition unit is above a threshold, based on the location information. The determination unit determines, based on the vehicle speed information, whether the speed distribution of the vehicle on the predetermined direction side at the position detected by the detection unit exhibits a bimodal pattern. as well as The sensing unit does not detect an obstacle at the location detected by the detection unit when the determination unit determines that the distribution exhibits bimodality, and detects an obstacle at the location detected by the detection unit when the determination unit determines that the distribution does not exhibit bimodality.

Images