Collision prediction device

Abstract

The present invention provides a collision prediction device that can improve collision prediction with a target traveling alongside during a turn. [Solution] The device 1 predicts a collision between a moving body CR and a target TR detected by sensors 11 and 12 mounted on the moving body, and includes a path prediction unit 110 that predicts a first movement path RT1 which is the movement path of the moving body and a second movement path RT2 which is the movement path of the target, a determination unit 120 that determines whether the moving body and the target are moving parallel to each other when the moving body turns, and a collision prediction unit 130 that uses the predicted first movement path and second movement path to predict whether or not there will be a collision between the moving body and the target, and predicts that the moving body and the target will not collide if it is determined that the moving body and the target are moving parallel to each other.

Description

Technical Field

[0001] The present disclosure relates to an apparatus for predicting collisions.

Background Art

[0002] Various techniques for predicting vehicle collisions have been proposed. For example, Patent Document 1 discloses a technique for estimating the respective movement paths of a host vehicle and targets around the host vehicle. In this technique, when the movement paths intersect, it is estimated that the host vehicle and the target will collide. The movement path of the host vehicle is estimated using the current position, speed, and acceleration of the host vehicle, and the movement path of the target is estimated using the current position and speed of the target.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the technique of Patent Document 1, there is a risk that the prediction accuracy of a collision may decrease due to an incorrect estimation of the movement path of a target running parallel to the host vehicle during a turn. Specifically, since the movement path of the target is estimated from the current position and speed, there are cases where, although the target is actually turning while running parallel to the host vehicle, it is estimated that the target is moving straight ahead. As a result, there is a risk of erroneously estimating that the host vehicle and the target will collide. Therefore, there is room for improvement in predicting a collision with a target running parallel to the host vehicle during a turn.

Means for Solving the Problems

[0005] The present disclosure can be realized in the following forms.

[0006] According to one embodiment of the present disclosure, a device is provided for predicting a collision between a moving body and a target detected by a sensor mounted on the moving body. The device includes: a path prediction unit that predicts a first movement path which is the movement path of the moving body and a second movement path which is the movement path of the target; a determination unit that determines whether the moving body and the target are moving parallel to each other when the moving body is turning; and a collision prediction unit that uses the predicted first movement path and the second movement path to predict whether or not a collision will occur between the moving body and the target, and predicts that the moving body and the target will not collide if it is determined that the moving body and the target are moving parallel to each other.

[0007] With this type of device, the collision prediction unit predicts that the moving object and the target will not collide when it determines that they are moving parallel to each other. This prevents the device from mistakenly predicting a collision between the moving object and the target when they are turning while moving parallel to each other. [Brief explanation of the drawing]

[0008] [Figure 1] This block diagram shows a schematic configuration of a collision prediction device in one embodiment of the present disclosure. [Figure 2] This diagram illustrates a situation where one vehicle and another vehicle are running parallel to each other. [Figure 3] This diagram illustrates the distance in the direction of the turning radius. [Figure 4] This is a flowchart showing the procedure for collision prediction processing during turning. [Modes for carrying out the invention]

[0009] A. Embodiments: <Overview of Collision Prediction Device 1> The collision prediction device 1 shown in Figure 1 is used when mounted on a vehicle. The collision prediction device 1 uses the detection results of various sensors mounted on the vehicle to predict whether or not there will be a collision with an object in the vicinity of the vehicle. In the following, the vehicle on which the collision prediction device 1 is mounted will also be referred to as the vehicle itself.

[0010] In this disclosure, "target" refers to the objects detected by various sensors mounted on a vehicle. Targets include, for example, pedestrians, bicycles, animals, other vehicles besides the vehicle itself, buildings, telephone poles, trees, etc. Among these targets, targets other than real estate that can move are referred to as "moving targets."

[0011] <Sensor Configuration> As shown in Figure 1, the collision prediction device 1 receives signals from multiple sensors via an input / output interface (I / F) 300. The multiple sensors shown in Figure 1 are mounted on the vehicle on which the collision prediction device 1 is installed. The multiple sensors include a front sensor 11, a side sensor 12, a vehicle speed sensor 13, a yaw rate sensor 14, and a steering angle sensor 15.

[0012] The front sensor 11 detects moving targets in the area including the front of the vehicle. The side sensor 12 detects moving targets in the area including the sides of the vehicle. Note that "sides of the vehicle" includes the front and rear sides of the vehicle. The front and rear sides refer to an axis along the longitudinal direction (forward / reverse direction) of the vehicle, and mean a range of 30 to 60 degrees from the axis passing through the center of the vehicle in the left-right direction. The front sensor 11 and the side sensor 12 are, for example, cameras and rangefinders. The rangefinders are, for example, devices that apply millimeter-wave radar or LiDAR (Light Detection and Ranging). The front sensor 11 is, for example, installed in the center of the front bumper of the vehicle. The side sensors 12 are, for example, installed at the left and right ends of the front bumper of the vehicle. The detection areas of the front sensor 11 and the side sensor 12 may partially overlap.

[0013] The vehicle speed sensor 13 generates a signal corresponding to the vehicle's speed. The vehicle speed sensor 13 detects, for example, the rotation speed of the vehicle's tires. The signal emitted by the vehicle speed sensor 13 is used by the collision prediction device 1 to calculate the vehicle's speed.

[0014] The yaw rate sensor 14 generates a signal corresponding to the yaw rate acting on the host vehicle. The signal generated by the yaw rate sensor 14 is used for the calculation of the yaw rate of the host vehicle by the collision prediction device 1.

[0015] The steering angle sensor 15 transmits a steering angle signal corresponding to the steering angle of the steering of the host vehicle to the collision prediction device 1. The steering angle sensor 15 is attached to the steering rod of the host vehicle.

[0016] <Configuration of the braking device 400> The braking device 400 applies a braking force for braking the host vehicle to the host vehicle. The braking device 400 applies the braking force according to the instruction of the occupant of the host vehicle or the instruction of the collision prediction device 1. The instruction of the occupant is given using an input device such as a brake pedal. The instruction of the collision prediction device 1 is given by transmission of an electric signal.

[0017] <Configuration of the collision prediction device 1> The collision prediction device 1 includes a processor 100, a memory 200, and an input / output I / F 300. Note that the collision prediction device 1 may be configured as a part of an ECU (Electronic Control Unit) that executes various controls of the host vehicle.

[0018] The processor 100 functions as a route prediction unit 110, a determination unit 120, and a collision prediction unit 130 by executing a program stored in the memory 200.

[0019] <Function of the route prediction unit 110> The route prediction unit 110 predicts a first movement route that is the movement route of the host vehicle and a second movement route that is the movement route of the target. Hereinafter, an example in which the target is another vehicle in motion will be described. As shown in FIG. 2, the host vehicle CR is traveling while turning on a curved road RD. The road RD has two lanes. The first movement route RT1 is indicated by a solid line. Another vehicle TR is traveling parallel to the host vehicle CR in the same direction while turning on the road RD. Another vehicle TR is traveling in a lane adjacent to the lane in which the host vehicle CR is traveling. The second movement route RT2 is indicated by a dashed line.

[0020] The route prediction unit 110 predicts the first movement route RT1 using the traveling speed of the host vehicle CR, the yaw rate, and the steering angle. The traveling speed of the host vehicle CR is calculated using the signal transmitted from the vehicle speed sensor 13 shown in FIG. 1. The yaw rate is calculated using the signal transmitted from the yaw rate sensor 14. The steering angle is calculated using the signal transmitted from the steering angle sensor 15.

[0021] The route prediction unit 110 predicts the second movement route RT2 using the detection results of the front sensor 11 and the side sensor 12. The route prediction unit 110 predicts the second movement route RT2 based on, for example, the change in the position of another vehicle TR detected by the front sensor 11 and the side sensor 12.

[0022] Furthermore, the path prediction unit 110 predicts the turning radius of the vehicle CR during a turn and predicts the distance (distance Xr, described later) between the vehicle CR and the other vehicle TR in the direction of the turning radius. The prediction of this distance will be explained using Figure 3. Figure 3 shows the vehicle CR as seen from above as it turns to the right. In Figure 3, the other vehicle TR is shown in a simplified form for illustrative purposes. As shown in Figure 3, the distance Xr can be calculated geometrically using the turning radius R of the vehicle CR, the lateral distance x and longitudinal distance y between the vehicle CR and the other vehicle TR, and the distance z from the sensor position to the rear axle. The turning radius R is calculated using the vehicle CR's travel speed and yaw rate. The lateral distance x and longitudinal distance y are calculated using the detection results of the front sensor 11 and the side sensor 12. The distance z from the sensor position to the rear axle is stored in memory 200 beforehand. By substituting these values ​​into equation (1) shown in Figure 3, the distance Xr is calculated. The route prediction unit 110 stores the calculated distance Xr in the memory 200 along with information about the time when the distance Xr was calculated.

[0023] <Functions of the determination unit 120> The determination unit 120 shown in Figure 1 determines whether the vehicle CR and the other vehicle TR will be traveling parallel to each other when the vehicle CR is turning, using the distance Xr predicted by the path prediction unit 110. More specifically, the determination unit 120 determines that the vehicle CR and the other vehicle TR will be traveling parallel if the amount of change in distance Xr over a predetermined period is less than a predetermined threshold, and determines that the vehicle CR and the other vehicle TR will not be traveling parallel if the amount of change over a predetermined period is greater than or equal to the threshold. In this embodiment, the predetermined period is 1 second. Note that the predetermined period is not limited to 1 second and may be set to any value. In this embodiment, the predetermined threshold is, for example, 50 cm. Note that the predetermined threshold is not limited to 50 cm and may be set to any value. The predetermined period and the predetermined threshold are stored in the memory 200. The determination unit 120 calculates the amount of change in distance Xr using the distance Xr stored in the memory 200 along with the time information. More specifically, the change in distance Xr is calculated using the latest distance Xr calculated by the route prediction unit 110 and the distance Xr at a predetermined time when the difference between the latest distance Xr and the time when the latest distance Xr was obtained is a predetermined period. Next, the determination unit 120 determines whether the change is less than a threshold. If the distance Xr is less than the threshold, that is, if the change in distance Xr is relatively small, the determination unit 120 determines that the vehicle CR and the other vehicle TR will be traveling in parallel. If the distance Xr is greater than or equal to the threshold, that is, if the change in distance Xr is relatively large, the determination unit 120 determines that the vehicle CR and the other vehicle TR will not be traveling in parallel.

[0024] <Functions of the collision prediction unit 130> The collision prediction unit 130 uses the predicted first movement path RT1 and second movement path RT2 to predict whether or not a collision will occur between the vehicle CR and the other vehicle TR. Specifically, the collision prediction unit 130 predicts whether or not a collision will occur by determining whether or not the first movement path RT1 and the second movement path RT2 intersect. If the collision prediction unit 130 predicts that the vehicle CR and the other vehicle TR will collide, it transmits a signal to the braking device 400 to apply braking force to the vehicle CR. This suppresses a collision between the vehicle CR and the other vehicle TR.

[0025] Furthermore, the collision prediction unit 130 in this disclosure predicts that the vehicle CR and the other vehicle TR will not collide, regardless of whether the first movement path RT1 and the second movement path RT2 intersect, if it is determined that the vehicle CR and the other vehicle TR are traveling side by side. In other words, even if the first movement path RT1 and the second movement path RT2 intersect, if it is determined that the vehicle CR and the other vehicle TR are traveling side by side, it predicts that the vehicle CR and the other vehicle TR will not collide. This prevents the system from mistakenly predicting a collision between the vehicle CR and the other vehicle TR when they are traveling side by side.

[0026] <Collision prediction processing during turning> The collision prediction process shown in Figure 4 is executed when the vehicle CR begins to turn. The collision prediction process is executed repeatedly while the vehicle CR is turning. The side sensor 12 detects another vehicle TR (step S110). Hereafter, "step S" will be simply referred to as "S".

[0027] The path prediction unit 110 predicts the first movement path RT1 of the vehicle CR and the second movement path RT2 of the other vehicle TR (S120). The path prediction unit 110 also predicts the distance Xr in the turning radius direction between the vehicle CR and the other vehicle TR (S130). The predicted distance Xr is written to the memory 200. Note that the processes of S120 and S130 may be executed in parallel.

[0028] The determination unit 120 determines whether a distance Xr predicted prior to the current time is stored in the memory 200 (S140). If the distance Xr is not stored in the memory 200 (S140: NO), that is, if the distance Xr to another vehicle TR is estimated for the first time, the process of S110 is executed again. If the distance Xr is stored in the memory 200 (S140: YES), that is, if the distance Xr at a time prior to the current time has already been estimated and stored, the determination unit 120 determines whether the vehicle CR and the other vehicle TR are running parallel to each other (S150). More specifically, the determination unit 120 calculates the change in distance Xr using the latest distance Xr estimated in S130 and the distance Xr at a previous time stored in the memory 200. The distance Xr at a previous time used here is predicted at a time when the period between that time and the time when the latest distance Xr was predicted is a predetermined period. The determination unit 120 determines that the vehicle CR and the other vehicle TR are running parallel if the calculated change amount is less than a predetermined threshold, and determines that the vehicle CR and the other vehicle TR are not running parallel if the calculated change amount is equal to or greater than the threshold.

[0029] If it is determined that the vehicle CR and the other vehicle TR are traveling side by side (S150: YES), the route prediction unit 110 predicts that the vehicle CR and the other vehicle TR will not collide (S160). Following S160, the process of S110 is executed again.

[0030] If it is determined that the vehicle CR and the other vehicle TR are not running parallel to each other (S150: NO), the collision prediction unit 130 predicts whether or not the vehicle CR and the other vehicle TR will collide (S170). If it is predicted that the vehicle CR and the other vehicle TR will collide (S170: YES), the collision prediction unit 130 transmits a control signal to the braking device 400 to increase the braking force (S180). If it is predicted that the vehicle CR and the other vehicle TR will not collide (S170: NO), the process of S110 is executed again.

[0031] According to the collision prediction device 1 of the embodiment described above, when the collision prediction unit 130 determines that its own vehicle CR and the other vehicle TR are traveling side by side, it predicts that the own vehicle CR and the other vehicle TR will not collide. Therefore, it is possible to suppress the erroneous prediction that the own vehicle CR and the other vehicle TR will collide when they are turning while traveling side by side.

[0032] Furthermore, according to the collision prediction device 1 of the embodiment, the determination unit 120 uses the distance Xr between the vehicle CR and the other vehicle TR in the turning radius direction, predicted by the path prediction unit 110, to determine whether the vehicle CR and the other vehicle TR are traveling parallel to each other. Compared to a configuration that uses a simple straight-line distance between the vehicle CR and the other vehicle TR, the determination can be performed using a more accurate positional relationship between the vehicle CR and the other vehicle TR during a turn. This suppresses the occurrence of misdetermination.

[0033] Furthermore, according to the collision prediction device 1 of the embodiment, the determination unit 120 determines that the vehicle CR and the other vehicle TR are driving side by side if the amount of change in distance Xr over a predetermined period is less than a preset threshold, and determines that the vehicle CR and the other vehicle TR are not driving side by side if the amount of change in distance Xr is equal to or greater than a preset threshold. Therefore, by setting an appropriate threshold, the determination of whether vehicles are driving side by side can be performed with high accuracy.

[0034] B. Other embodiments: (B1) In the above embodiment, the determination unit 120 may calculate the change in distance Xr using the distance Xr at an arbitrary first time point and the distance Xr at a second time point which is later than the first time point. The first time point is, for example, the timing when the vehicle CR begins to turn. The first time point is also, for example, the timing when the other vehicle TR is first detected. Furthermore, the determination unit 120 may set the threshold higher when the distance Xr at the first time point is larger. With this configuration, when the distance Xr is relatively large and the possibility of collision between the vehicle CR and the other vehicle TR is relatively small, a margin can be added to the threshold. This reduces the number of times the collision prediction unit 130 performs the collision prediction calculation between the vehicle CR and the other vehicle TR.

[0035] (B2) In the above embodiment, the determination unit 120 determined whether the vehicle CR and the other vehicle TR were traveling side by side using the distance Xr, but the disclosure is not limited thereto. The determination unit 120 may, for example, use the straight-line distance between the vehicle CR and the other vehicle TR to determine whether they are traveling side by side. In such a configuration, the determination unit 120 determines that the vehicles are traveling side by side if the amount of change in the straight-line distance over a predetermined time is less than a threshold, and determines that they are not traveling side by side if the amount of change in the straight-line distance over a predetermined time is equal to or greater than the threshold. With such a configuration, it is also possible to determine whether the vehicle CR and the other vehicle TR are traveling side by side.

[0036] (B3) In the above embodiment, the target traveling alongside the vehicle CR was another vehicle TR, but the disclosure is not limited thereto. The target traveling alongside the vehicle CR may be any moving object. Also, the collision prediction device 1 was mounted on a vehicle, but the disclosure is not limited thereto. The collision prediction device 1 may be mounted on any moving object. Examples of moving objects include ships, airplanes, so-called flying cars, etc.

[0037] (B4) In the above embodiment, the determination unit 120 determined that the vehicle CR and the other vehicle TR were running parallel if the amount of change in distance Xr over a predetermined period was less than a preset threshold, and determined that the vehicle CR and the other vehicle TR were not running parallel if the amount of change in distance Xr over a predetermined period was greater than or equal to a preset threshold. However, the disclosure is not limited thereto. The determination unit 120 may also determine that the vehicle CR and the other vehicle TR are running parallel if the amount of change in distance Xr over a predetermined period is less than or equal to a preset threshold, and determine that the vehicle CR and the other vehicle TR are not running parallel if the amount of change in distance Xr over a predetermined period is greater than a preset threshold.

[0038] (B5) The collision prediction device 1 and the methods described herein may be implemented by a dedicated computer provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. Alternatively, the collision prediction device 1 and the methods described herein may be implemented by a dedicated computer provided by configuring a processor by one or more dedicated hardware logic circuits. Alternatively, the collision prediction device 1 and the methods described herein may be implemented by one or more dedicated computers configured by a combination of a processor and memory programmed to perform one or more functions and a processor configured by one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by the computer on a computer-readable non-transitional tangible recording medium.

[0039] This disclosure is not limited to the embodiments described above, and can be implemented in various configurations without departing from its spirit. For example, the technical features in each embodiment corresponding to the technical features in the embodiments described in the summary of the invention can be replaced or combined as appropriate in order to solve some or all of the above-described problems, or to achieve some or all of the above-described effects. Furthermore, if a technical feature is not described as essential in this specification, it can be deleted as appropriate. This disclosure can be implemented, for example, in the form of a collision prediction method, a computer program for implementing such a method, or a non-temporary recording medium on which such computer program is recorded.

[0040] This disclosure may be implemented, for example, in the following forms: [Form 1] A device for predicting a collision between a moving object and a target detected by a sensor mounted on the moving object, A path prediction unit predicts a first movement path, which is the movement path of the moving body, and a second movement path, which is the movement path of the target. A determination unit that determines whether the moving body and the target move in parallel when the moving body rotates, A prediction unit that uses the predicted first movement path and the second movement path to predict whether or not a collision will occur between the moving body and the target, and a collision prediction unit that, when it is determined that the moving body and the target are moving parallel to each other, predicts that the moving body and the target will not collide; A device equipped with the following features. [Form 2] The apparatus described in Embodiment 1, The aforementioned path prediction unit, When the moving body rotates, the rotation radius of the moving body is predicted, Predict the distance between the moving body and the target in the direction of the predicted turning radius, The determination unit uses the distance to determine whether the moving object and the target are moving in parallel. Device. [Form 3] The apparatus described in Embodiment 2, The determination unit, If the amount of change in the aforementioned distance over a predetermined period is less than a predetermined threshold, it is determined that the moving object and the target are moving in parallel. If the amount of change in the distance over the period is greater than or equal to the threshold, it is determined that the moving body and the target are not moving in parallel. Device. [Form 4] The apparatus described in Embodiment 3, The determination unit, Using the distance at the first time point and the distance at the second time point, which is a later time point than the first time point, the amount of change in the distance is calculated. If the distance at the first time point is greater, the threshold is set to be larger. Device. [Explanation of Symbols]

[0041] 1...Collision prediction device, 11...Forward sensor, 12...Side sensor, 13...Vehicle speed sensor, 14...Yaw rate sensor, 15...Steering angle sensor, 100...Processor, 110...Path prediction unit, 120...Determination unit, 130...Collision prediction unit, 200...Memory, 300...Input / output interface, 400...Braking device, CR...Own vehicle, R...Turning radius, RD...Road, RT1...First movement path, RT2...Second movement path, TR...Other vehicle, Xr...Distance in the turning radius direction, x...Lateral distance, y...Longitudinal distance, z...Distance from sensor position to rear axle

Claims

1. A device (1) for predicting a collision between a moving body (CR) and a target (TR) detected by sensors (11, 12) mounted on the moving body, A path prediction unit (110) predicts a first movement path (RT1), which is the movement path of the moving body, and a second movement path (RT2), which is the movement path of the target. A determination unit (120) that determines whether the moving body and the target move in parallel when the moving body rotates, A prediction unit that uses the predicted first movement path and the second movement path to predict whether or not a collision will occur between the moving body and the target, and a collision prediction unit (130) that determines that the moving body and the target will move in parallel and predicts that the moving body and the target will not collide, A device equipped with the following features.

2. The apparatus according to claim 1, The aforementioned path prediction unit, When the moving body rotates, the rotation radius of the moving body is predicted, The distance (Rx) between the moving body and the target in the direction of the predicted turning radius is predicted, The determination unit uses the distance to determine whether the moving object and the target are moving in parallel. Device.

3. The apparatus according to claim 2, The determination unit, If the amount of change in the aforementioned distance over a predetermined period is less than a predetermined threshold, it is determined that the moving object and the target are moving in parallel. If the amount of change in the distance over the period is greater than or equal to the threshold, it is determined that the moving body and the target are not moving in parallel. Device.

4. The apparatus according to claim 3, The determination unit, Using the distance at the first time point and the distance at the second time point, which is a time point later than the first time point, the amount of change in the distance is calculated. If the distance at the first time point is greater, the threshold is set to be larger. Device.

Images