Driving assistance system and driving assistance method

Abstract

The present invention comprises: an object information acquisition unit that acquires an object detection distance, an object line-of-sight speed, and an object detection angle for an object detected by an object recognition sensor mounted on a host vehicle; a vehicle speed information acquisition unit that acquires a host vehicle speed, which is the speed of the host vehicle; a tracking processing unit that tracks the object on the basis of the distance, the object line-of-sight speed, and the object detection angle to acquire the movement trajectory of the object; a relative angle calculation unit that calculates a relative angle, which is the angle at which the object is detected with respect to the travel direction of the host vehicle, from the object detection angle and an attachment angle at which the object recognition sensor is attached to the host vehicle; and an assessment unit that assesses whether the tracked object is a stationary object or a moving object on the basis of a ground line-of-sight speed, which is the sum of the host vehicle speed, the cosine multiplication value of the relative angle, and the object line-of-sight speed for the object acquired by the tracking processing unit.

Description

Driving assistance system and driving assistance method

[0001] The present invention relates to a driving assistance system that distinguishes between stationary and moving objects.

[0002] In recent years, vehicles have been required to improve their collision safety performance against pedestrians and cyclists. In particular, since the Convention Regulations are applied as laws in each country, the installation of collision safety devices is mandatory. Convention Regulation No. 151, among other requirements, requires that an alarm be activated even when the relative speed between the vehicle and a detected object is 0 m / s. To achieve this function, millimeter-wave radar is sometimes used to measure the distance and speed of a detected object. However, since the line-of-sight velocity, which is the speed at which an object approaches the millimeter-wave radar, cannot extract the circumferential velocity component, it is difficult to distinguish between a detected object traveling parallel to the sensor and a wall located directly to the sensor.

[0003] Conventionally, if a target equivalent to a planar wall is detected by a large number of detection points, it is recognized as a wall, and if there are only a few detection points, it is recognized as a parallel running object.

[0004] The following prior art exists as background art in this technical field: Patent Document 1 (JP 2007-292643 A) describes a peripheral object identification device that is mounted on an automobile and identifies objects around the automobile, the peripheral object identification device including a radar device that detects the relative positions of objects around the automobile relative to the automobile, and a roadside fixed object identification means that, when the radar device detects an object to the side of the automobile whose relative distance to the automobile does not change, identifies the detected object as an identification target object and identifies whether the identification target object is a roadside fixed object that is an object fixed to the roadside, the roadside fixed object identification means determining whether another object is present in a position that forms a line with the identification target object based on the relative positions of each object detected by the radar device, and if the other object is present, identifies the identification target object as a roadside fixed object.

[0005] Japanese Patent Application Laid-Open No. 2007-292643

[0006] Patent Document 1 uses information different from that detected by millimeter-wave radar to determine whether an object is traveling parallel to the vehicle. In other words, in a situation where there is a wall, guardrail, etc. directly to the side of the sensor, tracking data at the same lateral position as the tracking data directly to the side is extracted, and the relative speed of the extracted tracking data is the vehicle speed × -1 to distinguish between a wall and a parallel object.

[0007] In Patent Document 1, it is assumed that the detection point changes in accordance with the amount of movement of the vehicle when a protrusion is present on the wall. Therefore, when there is no protrusion on the wall, the position of the detection point does not change, making it difficult to apply the method described in Patent Document 1.

[0008] The present invention has been made in consideration of the above-mentioned problems, and provides a method for determining stationary objects even in situations where the position of the detection point relative to the wall does not change, by using a sensor characteristic that can detect line-of-sight velocity using Doppler frequency.

[0009] A representative example of the invention disclosed in the present application is as follows: That is, the vehicle information acquisition unit acquires, for an object detected by an object recognition sensor mounted on a host vehicle, an object detection distance indicating the distance between the object recognition sensor and the detected object, an object line-of-sight velocity which is the relative velocity of the detected object in the detection direction, and an object detection angle which is the direction in which the object was detected, a vehicle speed information acquisition unit acquires the host vehicle speed which is the speed of the host vehicle, a tracking processing unit tracks the object and acquires a movement trajectory of the object based on the distance, the object line-of-sight velocity, and the object detection angle, a relative angle calculation unit calculates, from the object detection angle and the mounting angle of the object recognition sensor on the host vehicle, a relative angle which is the angle at which the object was detected with respect to the traveling direction of the host vehicle, and a determination unit which determines whether the tracked object is a stationary object or a moving object based on the ground line-of-sight velocity which is the sum of the product of the host vehicle speed and the cosine of the relative angle for the object acquired by the tracking processing unit and the object line-of-sight velocity.

[0010] According to one aspect of the present invention, it is possible to distinguish between stationary and moving objects in a situation where the position of the detection point relative to the wall does not change, multiple points cannot be detected, and it is impossible to distinguish between a wall and a moving object. Problems, configurations, and effects other than those described above will become clear from the description of the following embodiments.

[0011] FIG. 1 is a functional block diagram of a driving assistance system according to an embodiment of the present invention. FIG. 2 is a diagram showing an environment in which it is difficult to recognize stationary objects in embodiment 1 of the present invention. FIG. 3 is a flowchart of processing executed by a stationary / moving object determination unit according to embodiment 1 of the present invention. FIG. 4 is a diagram showing wall determination in step S03 according to embodiment 1 of the present invention. FIG. 5 is a diagram showing results of processing by the stationary / moving object determination unit according to embodiment 1 of the present invention. FIG. 6 is a flowchart of processing executed by a stationary / moving object determination unit according to embodiment 2 of the present invention. FIG. 7 is a flowchart of processing executed by a stationary / moving object determination unit according to embodiment 3 of the present invention.

[0012] First Embodiment A vehicle control device (ECU) 10 according to a first embodiment will be described below with reference to FIGS. 1 to 5. FIG.

[0013] FIG. 1 is a functional block diagram of a driving assistance system according to this embodiment.

[0014] The driving assistance system of this embodiment is mounted on the vehicle 1, and an object recognition sensor 21, a vehicle speed sensor 22, a steering sensor 23, and a yaw rate sensor 24 are connected to the input side of a vehicle control device 10, and an alarm device 31 is connected to the output side of the vehicle control device 10. The sensors and devices are connected to each other so that they can communicate with each other via a controller area network (CAN) or the like.

[0015] The object recognition sensor 21 is a sensor that acquires information about an object 3 (e.g., another vehicle, a pedestrian) around the vehicle 1, and is, for example, a millimeter wave radar 21a, an ultrasonic sensor, a LiDAR, etc. If the object recognition sensor 21 is a millimeter wave radar 21a, the relative position P TR (x TR , y TR ) is directly measured, and the speed at which the object 2 approaches the vehicle 1 (i.e., the relative speed in the detection direction) is the object line-of-sight speed V Rad can be detected from the Doppler frequency.

[0016] In this embodiment, the object recognition sensor 21 is a millimeter wave radar 21a, and the object line of sight velocity V Rad The control details for directly measuring the distance to the object, the direction of the object, and the line-of-sight velocity V Rad Any sensor that can measure the distance to an object and the direction of the object may be replaced with the relative position XY of the object.

[0017] The vehicle speed sensor 22 detects the vehicle speed V S For example, the wheel speed sensor is a sensor that detects the vehicle speed V on the assumption that the vehicle 1 is traveling straight, based on the rotation speed of the tire output from the wheel speed sensor. S Calculate.

[0018] The steering sensor 23 is a sensor that acquires the steering angle (operation amount) of the vehicle 1, and is, for example, an angle sensor attached to the steering device.

[0019] The yaw rate sensor 24 is a sensor that acquires the yaw rate ω of the host vehicle 1, and is, for example, an acceleration sensor that detects acceleration around the vertical axis of the host vehicle 1. Note that since the yaw rate ω can be calculated by a method described later, the yaw rate sensor 24 may be omitted.

[0020] The warning device 31 is a device that warns the driver of the possibility of a collision with the object 2 by displaying a message on a display, illuminating an LED, operating the steering wheel, issuing a voice notification, or the like, in accordance with a warning request output by the vehicle control device 10, and prompts the driver to take appropriate avoidance action. Note that, although the driver is prompted to take avoidance action via the warning device 31 in Fig. 1, the vehicle control device 10 may directly control the braking and steering of the vehicle 1 when there is a possibility of a collision.

[0021] <Vehicle Control Device 10> Next, the configuration of the driving assistance system of this embodiment will be described. As shown in FIG. 1 , the vehicle control device 10 includes a parameter storage unit 11, a tracking generation unit 12, a stationary / moving object determination unit 13, and an alarm system 14. The tracking generation unit 12, the stationary / moving object determination unit 13, and the alarm system 14 periodically execute processes, generate an alarm request when a predetermined condition is met, and output the generated alarm request to the alarm device 31. Specifically, the vehicle control device 10, specifically the data collection device A2, is an ECU (Electronic Control Unit) having an arithmetic unit, a storage device, and a communication interface. The arithmetic unit is a processor (e.g., a microcomputer) that executes a program stored in the storage device. By executing a predetermined program, the arithmetic unit operates as various functional blocks, such as the stationary / moving object determination unit 13, that provide various functions. The storage device includes a non-volatile storage area and a volatile storage area. The non-volatile storage area includes a program area accessible by the arithmetic unit and storing programs executed by the arithmetic unit, and a data area that temporarily stores data used by the arithmetic unit when executing the program. The volatile storage area stores data used by the computing device when the program is executed. The communication interface connects to other electronic control units via a network such as CAN or Ethernet.

[0022] The parameter storage unit 11 is a storage area for storing vehicle parameters and warning parameters. The vehicle parameters are parameters mainly used by the stationary / moving object determination unit 13, such as parameters related to the specifications of the vehicle 1, such as a circumferential movement threshold, a reliability threshold, a wheel base, and a gear ratio for converting a steering angle into a tire angle. The warning parameters are parameters mainly used by the warning unit 15, such as a Time To Collision (TTC) threshold for comparison with a warning candidate.

[0023] The tracking generation unit 12 uses information from the object recognition sensor 21, the vehicle speed sensor 22, etc. to track the target detected by the object recognition sensor 21 and estimates a tracking that represents the path of the target.

[0024] The stationary / moving object determination unit 13 determines whether the target is a stationary object based on the tracking estimated by the tracking generation unit 12. The stationary / moving object determination unit 13 may execute processing prior to the tracking generation unit 12. In the embodiment, the stationary / moving object determination unit 13 determines whether the target is a stationary object based on the tracking generated by the tracking generation unit 12 using the observation results of the millimeter wave radar 21a.

[0025] The driving assistance system of this embodiment may be applied to improve the detection accuracy of a device that generates a point cloud like a Lidar (for example, a millimeter-wave radar 21a), which has difficulty in separate detection like a camera. To recognize a single stationary object in the environment shown in FIG. 2 , it is necessary to widen the range of the clustering position difference and line-of-sight velocity difference. Widening the position difference and line-of-sight velocity difference makes it difficult to separate and detect a group of pedestrians. In order to improve the stationary object determination accuracy while maintaining the detection performance of pedestrians, etc., the system does not use a single tracking method, but instead uses the detected surrounding information in the process shown in FIG. 3 .

[0026] The warning system 14 calculates the possibility of a collision and a collision grace period using the target recognition result by the object recognition sensor 21, the vehicle speed detected by the vehicle speed sensor 22, the steering angle acquired by the steering sensor 23, the yaw rate acquired by the yaw rate sensor 24, and the vehicle parameters and warning parameters stored in the parameter storage unit 11, and outputs a warning to alert the driver when the possibility of a collision is high. The warning system 14 may output the warning using the warning device 31, or the ECU may output the warning. In addition to the warning, the warning system 14 may also output signals to control the vehicle, such as braking and steering.

[0027] Next, the processing executed by the stationary / moving object determination unit 13 will be described with reference to the flowchart of FIG.

[0028] Step S01: First, it is determined whether or not the object is moving based on the distance, line-of-sight velocity, angle, and vehicle speed detected by the vehicle speed sensor 22 of the detection data associated with tracking. For example, as shown in the following formula, the vehicle speed V S and the cosine of the detection angle θ, and the detected line of sight velocity V RradAdding these together, the ground velocity is V Arad Calculate. Formula 1

[0029]

[0030] The ground velocity V calculated using Equation 1 Arad is equal to or greater than the threshold, it is determined that the object is moving (S21). Arad If is less than the threshold value, the determination in S02 is performed.

[0031] Step S02: It is determined whether the tracking object that cannot be determined as a moving object in step S01 is a stationary object. The radial velocity output from the millimeter wave radar 21a is only the velocity coming towards the millimeter wave radar 21a. Therefore, the ground speed in the circumferential direction cannot be detected. In particular, in the direction perpendicular to the traveling direction of the vehicle, both moving objects and stationary objects traveling in the same direction as the vehicle are detected as 0 m / s. Therefore, the radial velocity V Arad If tracking is detected in a direction perpendicular to the traveling direction of the vehicle 1 where the ground velocity V is less than the threshold, the object is determined to be stationary since this can be determined from the line-of-sight velocity (S22). Arad If tracking below the threshold is detected in a direction that is not perpendicular to the traveling direction of the vehicle 1, it is not possible to determine whether a moving object whose direction of movement is perpendicular to the traveling direction of the vehicle is stationary or moving, and therefore a determination is made in step S03.

[0032] Step S03: For tracking objects that cannot be determined as stationary or moving in step S02, it is determined whether they are stationary or not based on the detection data around the tracking object. For tracking objects that cannot be determined as stationary in step S02, it is difficult to determine whether they are stationary or moving based on the detection data of the tracking object alone. In step S03, since it is possible that an object that is difficult to determine based on detection data alone is difficult to determine because a long object such as wall 2a was detected only from the side, as shown in Figure 4, it is estimated that whether the tracked object is stationary or moving is the same as the detection data that exists in the front and rear directions.

[0033] Detection data of object 3b, which has the same lateral position as the tracked object 3a and is located in the fore-and-aft direction of the tracked object 3a for which it cannot be determined whether it is a stationary or moving object in S02, is extracted. The detection data of object 3b extracted in the fore-and-aft direction is subjected to the processing of step S01 to determine whether it is a stationary or moving object. If object 3b is a stationary object, there is a high possibility that it will collide with the tracked object 3a located directly beside the vehicle 1, or that it is detection data separated from other tracking in the fore-and-aft direction. Therefore, if the detection data extracted in the fore-and-aft direction is a stationary object, the tracked object is determined to be a stationary object (S04). On the other hand, if the detection data extracted in the fore-and-aft direction is a moving object, or if detection data cannot be extracted, the tracking object located directly beside the vehicle 1 is determined to be a moving object (S23).

[0034] <Simulation Results> The effect of applying the present invention to a case where a stationary object is determined in a situation where the position of the detection point relative to the wall 2a does not change will be described. As shown in FIG. 2, when a vehicle is traveling at 10 km / h beside the wall 2a, objects are detected at the positions shown in FIGS. 5(A) and 5(B). In this case, the line-of-sight velocities of the tracked object 3a and the wall 2a are as shown in FIG. 5(C). In FIG. 5(C), the detection point where the line-of-sight velocity is approximately 0 is the detection result of the tracked object 3a, and the detection point where the line-of-sight velocity is far from 0 is the detection result of the detection data of the object 3b extracted in the forward and backward directions of the tracked object. In the processing of steps S01 and S03, the ground line-of-sight velocity is determined using a predetermined threshold value (+0.5 and −0.5 in the figures), and detected objects within the range defined by the threshold value are determined to be stationary objects (see FIG. 5(D)). Then, in the process of step S02, the angle is judged based on a predetermined threshold (90 degrees ±30 degrees in the figure), and a detected object is observed at the same lateral position that is larger than the range determined by the threshold (see FIG. 5(E)). Then, in the process of step S03, it is judged that the detection data existing in the forward and backward directions of the tracked object and the judgment result of whether the object is stationary or moving are the same (see FIG. 5(F)).

[0035] In this way, by the processing of this embodiment, an object that would otherwise be determined to be a moving object running in parallel can be determined to be a stationary object.

[0036] Therefore, according to this embodiment, a stationary object such as the wall 2a that continues in the traveling direction can be correctly recognized as a stationary object, and the recognition accuracy of the stationary object can be improved. This makes it possible to determine the possibility of a collision with an appropriate object and issue an alarm or control.

[0037] <Example 2> In Example 2, a situation is considered in which a moving object exists directly to the side and stationary objects exist in the front and rear directions, as shown in Fig. 6. In Example 2, differences from Example 1 will be mainly described, and descriptions of the same configurations and processes as Example 1 will be omitted. In Example 2, the process (Fig. 3) executed by the stationary object / moving object determination unit 13 differs in step S03, but the other configurations and processes are the same as those in Example 1 described above.

[0038] In step S03 of the second embodiment, detection data of the surroundings within a predetermined distance in the forward and backward directions from the object to be determined is extracted. The predetermined forward distance is set to a distance at which a moving object may collide with a stationary object, making avoidance action necessary. The predetermined rearward distance is set to a short distance because the distance between a moving object and a stationary object behind it increases over time.

[0039] In step S03 of Example 1, since the front-to-rear direction is not restricted, there is a possibility that a moving object directly to the side may be erroneously determined as a stationary object. However, by restricting the extraction range of the detection data in the front-to-rear direction in step S03 of Example 2, it is possible to suppress erroneous determination due to unexpected objects.

[0040] <Example 3> In Example 3, as shown in Fig. 6 , in a situation where a moving object is present directly to the side and a stationary object is present in the front and rear directions, the moving object directly to the side and the stationary object in the front and rear directions are determined to be different objects by a means different from that of Example 2. Note that in Example 3, differences from Example 1 will be mainly described, and descriptions of the same configurations and processes as Example 1 will be omitted. In Example 3, the process ( Fig. 3 ) executed by the stationary object / moving object determination unit 13 differs in step S03, and the other configurations and processes are the same as those of Example 1 described above.

[0041] When detecting a wall 2a, as shown in Figure 7, the millimeter-wave radar 21a detects the strongest reflected waves from directly beside the wall. For objects positioned further back or forward from directly beside the wall, the reflection angle becomes oblique, resulting in a smaller amount of reflection. Furthermore, since the distance is closest from directly beside the wall, the spatial propagation loss is also small. Therefore, the radio wave intensity detected from the wall 2a at a position further back or forward from directly beside the wall is smaller than the radio wave intensity detected from the wall 2a from directly beside the wall.

[0042] In step S03 of Example 3, the radio wave intensity of the detection data of object 3b located in the fore-and-aft direction of object 3a detected directly beside it (i.e., in the same lateral position) is compared with the detection data of object 3a detected directly beside it, and if the radio wave intensity of the reflected waves from a direction other than directly beside it is higher than the radio wave intensity of the reflected waves from directly beside it is highly likely that object 3b generating the reflected waves from a direction other than directly beside it is an object different from object 3a detected directly beside it, and the extracted data of object 3b is excluded from the stationary detection points.

[0043] The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the spirit and scope of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to configurations including all of the described configurations. Furthermore, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment may be added, deleted, or replaced with other configurations.

[0044] Furthermore, the aforementioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole in hardware, for example by designing them as integrated circuits, or may be realized in software by having a processor interpret and execute a program that realizes each function.

[0045] Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium such as an IC card, an SD card, or a DVD.

[0046] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and do not necessarily show all the control lines and information lines that are necessary for implementation. In reality, it can be considered that almost all components are interconnected.

Claims

1. A driving assistance system comprising: an object information acquisition unit that acquires, for an object detected by an object recognition sensor mounted on a host vehicle, an object detection distance that indicates the distance between the object recognition sensor and the detected object, an object line-of-sight velocity that is the relative speed in the detection direction of the detected object, and an object detection angle that indicates the direction in which the object was detected; a vehicle speed information acquisition unit that acquires the host vehicle speed that is the speed of the host vehicle; a tracking processing unit that tracks the object and acquires a movement trajectory of the object based on the distance, the object line-of-sight velocity, and the object detection angle; a relative angle calculation unit that calculates the relative angle that is the angle at which the object was detected with respect to the traveling direction of the host vehicle, from the object detection angle and the mounting angle of the object recognition sensor on the host vehicle; and a determination unit that determines whether the tracked object is a stationary object or a moving object based on the ground line-of-sight velocity that is the sum of the product of the host vehicle speed and the cosine of the relative angle for the object acquired by the tracking processing unit and the object line-of-sight velocity.

2. A driving assistance system according to claim 1, wherein the object line-of-sight velocity is a velocity detected based on a Doppler frequency.

3. A driving assistance system as described in claim 1, characterized in that the determination unit determines that the tracked object is a moving object when the magnitude of the ground line-of-sight velocity of the tracked object is outside a predetermined speed range.

4. A driving assistance system as described in claim 3, characterized in that the determination unit determines that the tracked object is a stationary object when the magnitude of the ground line-of-sight velocity of the tracked object is within the predetermined speed range and the magnitude of the relative angle is within a predetermined angle range.

5. A driving assistance system as described in claim 4, wherein the determination unit, when the magnitude of the ground line of sight velocity of the tracked object is within the predetermined speed range and the magnitude of the relative angle is outside the predetermined angle range, determines whether the tracked object is a stationary object or a moving object using the ground line of sight velocity of a second detected object different from the tracked object and located in the forward or backward direction of the tracked object, and in determining whether the tracked object is a stationary object or a moving object, when the ground line of sight velocity of the second detected object is within the predetermined speed range, determines the tracked object to be a stationary object, and when the ground line of sight velocity of the second detected object is outside the predetermined speed range, determines the tracked object to be a moving object.

6. A driving assistance system according to claim 5, wherein the second detected object is a detected object that exists within a predetermined range in front of and behind the tracked object.

7. A driving assistance system according to claim 5, characterized in that the strength of the signal reflected by the second detected object is lower than the strength of the signal reflected by the tracked object.

8. A driving assistance method executed by a driving assistance system, the driving assistance system having a calculation device that executes a program and a storage device accessible by the calculation device, the driving assistance method comprising: an object information acquisition procedure in which the calculation device acquires, for an object detected by an object recognition sensor mounted on the host vehicle, an object detection distance indicating the distance between the object recognition sensor and the detected object, an object line-of-sight speed that is the relative speed of the detected object in the detection direction, and an object detection angle that indicates the direction in which the object was detected; a vehicle speed information acquisition procedure in which the calculation device acquires the host vehicle speed that is the speed of the host vehicle; a tracking processing procedure in which the calculation device tracks the object based on the distance, the object line-of-sight speed, and the object detection angle, and acquires a movement trajectory of the object; and a relative angle calculation procedure in which the calculation device calculates the relative angle that is the angle at which the object was detected with respect to the traveling direction of the host vehicle, from the object detection angle and an attachment angle of the object recognition sensor to the host vehicle. a determination procedure in which the arithmetic device determines whether the tracked object is a stationary object or a moving object based on a ground line-of-sight velocity, which is the sum of the product of the host vehicle speed and the cosine of the relative angle for the object acquired in the tracking processing procedure and the object line-of-sight velocity.