A method for determining a lateral distance jump across a target

Abstract

The application provides a judgment method for jumping of a transverse distance of a target, and relates to the technical field of auxiliary driving, and comprises the following steps: obtaining target information through a collection device, wherein the target information comprises a transverse distance, a longitudinal distance, and a azimuth angle of the target in a vehicle coordinate system; obtaining a yaw rate of the vehicle; calculating a projection speed of a related speed of the target in the X direction of the vehicle coordinate system; calculating a maximum transverse motion relative speed of the target according to a preset maximum transverse motion absolute speed threshold of the target and the projection speed; calculating a maximum transverse distance change amount of the target between frames according to the maximum transverse motion relative speed of the target and a sampling frequency of the target information; calculating a transverse distance change amount of the target between a current frame and a previous frame; comparing the transverse distance change amount of the target with the maximum transverse distance change amount of the target between frames; and outputting a target transverse information confidence flag. The application significantly reduces the AEB false triggering rate of the transverse pedestrians and cyclists, and improves the system robustness.

Description

Technical Field

[0001] This invention relates to the field of driver assistance technology, and specifically to a method for determining changes in lateral distance when crossing a target. Background Technology

[0002] Automatic Emergency Braking (AEB) is a basic function of Advanced Driving Assistant Systems (ADAS). It can identify and analyze traffic conditions ahead in the direction of travel using a single camera or a combination of camera and radar, and will warn the driver or automatically brake to mitigate or avoid a collision when a collision is possible.

[0003] For target detection schemes that fuse cameras and radar, millimeter-wave radar has high accuracy in longitudinal velocity and distance measurement, ensuring accurate judgment of the target's longitudinal motion state in the fused state. However, since cameras have higher accuracy in lateral velocity and distance measurement than millimeter-wave radar, the accuracy of judging the target's lateral motion state in the fused state is primarily determined by the camera. However, cameras, as visual sensors, are significantly affected by external environmental interference. Detection of small targets such as pedestrians / cyclists may be affected by background objects or lighting conditions, leading to inaccurate or jittering target detection boxes, thus impacting velocity and distance measurement performance. Pedestrian / cyclist crossing is a typical scenario for AEB (Autonomous Emergency Braking). The accuracy of lateral distance measurement directly affects the accuracy of historical points and lateral velocity during the crossing process. If the lateral distance changes abruptly due to target detection box jitter, the system may misjudge the target's lateral motion state, further leading to the AEB system misjudging the collision risk and causing false triggering. Summary of the Invention

[0004] To address the problem of misjudgment by AEB systems due to target detection box jitter in existing technologies, this application provides a method for judging lateral distance jumps when crossing a target, thereby improving the accuracy of judging lateral distance jumps when crossing a target and thus reducing the false trigger rate of AEB for pedestrians and cyclists crossing targets.

[0005] This application provides the following technical solution: a method for determining lateral distance jumps when crossing a target, including: Target information is acquired through a data acquisition device, and the target information includes the target's lateral distance, longitudinal distance, and azimuth angle in the vehicle coordinate system. The yaw rate of the vehicle is obtained, and based on the yaw rate of the vehicle, the lateral distance, longitudinal distance, and azimuth angle, the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system is calculated. The maximum lateral relative velocity of the target is calculated based on the preset threshold of the absolute velocity of the target's maximum lateral movement and the projected velocity. Calculate the maximum change in the target's lateral distance between frames based on the target's maximum relative lateral velocity and the sampling frequency of the target information; Calculate the change in the target's lateral distance between the current frame and the previous frame, compare the change in the target's lateral distance with the maximum change in the target's lateral distance between frames, and output the target's lateral information confidence flag based on the comparison result.

[0006] According to one embodiment of this application, the projected velocity of the target's entrainment velocity in the X direction of the vehicle coordinate system is calculated by the following formula:

[0007] in, ω represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system, ω represents the yaw rate of the vehicle, x represents the lateral distance of the target in the vehicle's coordinate system, y represents the longitudinal distance of the target in the vehicle's coordinate system, and θ represents the azimuth angle of the target in the vehicle's coordinate system.

[0008] According to one embodiment of this application, if the acquisition device can obtain the orientation information of the target, the maximum lateral relative velocity of the target is calculated using the following formula:

[0009] in, This represents the target's maximum lateral relative velocity. This represents the preset threshold for the maximum absolute lateral velocity of the target. This represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system.

[0010] According to one embodiment of this application, if the acquisition device cannot obtain the orientation information of the target, the maximum lateral relative velocity of the target is calculated using the following formula:

[0011] in, This represents the target's maximum lateral relative velocity. This represents the preset threshold for the maximum absolute lateral velocity of the target. This represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system.

[0012] According to one embodiment of this application, the maximum change in the inter-frame target lateral distance is calculated using the following formula:

[0013] in, This represents the maximum change in the lateral distance between targets between frames. denoted by , where v represents the target's maximum lateral relative velocity, and v represents the sampling frequency of the target information.

[0014] According to one embodiment of this application, the function f(x, of the target lateral information confidence flag bit) is... The definition is: if |x(k) - x(k-1)|> If f=0, it means the horizontal information is unreliable; if |x(k) - x(k-1)|≤ If f=1, it indicates that the horizontal information is reliable; in, Indicates the horizontal distance of the target in the current frame. This indicates the horizontal distance of the target in the previous frame. This indicates the maximum change in the horizontal distance between targets in different frames.

[0015] According to one embodiment of this application, the yaw rate of the vehicle is obtained via the CAN bus of the vehicle body.

[0016] Compared with the prior art, the beneficial effects that can be achieved by the above-mentioned at least one technical solution adopted in the embodiments of this specification include at least the following: By introducing entanglement speed compensation and physically reasonable threshold judgment, the embodiments of this invention can effectively identify and mark unreliable lateral distance jumps caused by perceived noise, providing key status information for the collision risk judgment module of the AEB system, thereby significantly reducing the false trigger rate of AEB for pedestrians and cyclists crossing the road and improving the robustness of the system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of the method for determining the lateral distance jump when crossing a target according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the vehicle coordinate system during the calculation process of an embodiment of the present invention. Detailed Implementation

[0019] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0020] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] like Figure 1 As shown, this embodiment of the invention provides a method for determining a jump in lateral distance when crossing a target, including the following steps: Step 1: Acquire target information using a data acquisition device. The target information includes the target's lateral distance, longitudinal distance, and azimuth angle in the vehicle coordinate system. Step 2: Obtain the yaw rate of the vehicle. Based on the yaw rate of the vehicle, the lateral distance, longitudinal distance, and azimuth angle, calculate the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system. Step 3: Calculate the target's maximum lateral relative velocity based on the preset target maximum absolute lateral velocity threshold and the projected velocity; Step 4: Calculate the maximum change in the target's lateral distance between frames based on the target's maximum relative lateral velocity and the sampling frequency of the target information; Step 5: Calculate the change in the target's lateral distance between the current frame and the previous frame, compare the change in the target's lateral distance with the maximum change in the target's lateral distance between frames, and output the target's lateral information confidence flag based on the comparison result.

[0022] The method for determining lateral distance jumps when crossing targets according to embodiments of the present invention obtains the lateral distance, longitudinal distance, and azimuth angle of the target in the vehicle coordinate system, and obtains the yaw rate of the vehicle; calculates the projection of the target's entrainment velocity in the lateral direction of the vehicle coordinate system; calculates the target's maximum lateral relative velocity based on a set threshold for the target's maximum absolute lateral velocity; further calculates the maximum change in the target's lateral distance between frames; and determines whether a lateral distance jump has occurred by comparing the actual change in lateral distance between frames with the maximum change in the target's lateral distance between frames, and outputs a confidence flag. This invention can effectively identify lateral distance jumps caused by target detection box jitter, providing reliable lateral information confidence judgment for the AEB system, thereby reducing the false trigger rate for pedestrians, cyclists, and other targets crossing targets.

[0023] In practice, the specific steps of the method in this embodiment are as follows: Step 1: Data Acquisition The vehicle's lateral distance (x), longitudinal distance (y), and azimuth angle (θ) in the vehicle's coordinate system are acquired using an onboard camera (counter-clockwise direction is defined as positive). Simultaneously, if the camera supports this, the target's orientation information is also acquired. The vehicle's yaw rate (ω) is acquired in real-time via the vehicle's CAN bus (counter-clockwise direction is defined as positive). The sampling frequency for the target information is set to v.

[0024] Step 2: Calculate the projection of the entrainment velocity like Figure 2 As shown, due to the vehicle's motion, the target will have a drag velocity. Calculate the projection of this drag velocity onto the X-axis (horizontal) of the vehicle's coordinate system. The calculation formula is:

[0025] in, ω represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system, ω represents the yaw rate of the vehicle, x represents the lateral distance of the target in the vehicle's coordinate system, y represents the longitudinal distance of the target in the vehicle's coordinate system, and θ represents the azimuth angle of the target in the vehicle's coordinate system.

[0026] Since the vehicle's coordinate system is a moving system when it is in motion, the target will have an entrainment velocity. The target entrainment velocity refers to the velocity of the point in the vehicle's coordinate system that coincides with the target (i.e., the entrainment point) relative to the Earth's coordinate system.

[0027] Because in vector relations, the absolute velocity V a Relative velocity V r and the speed of entrapment V t The following relationship is satisfied: Va=Vr+Vt. Therefore, this step calculates the projection of the entrainment velocity in the lateral direction, which is a necessary input for the next step to calculate the target's maximum lateral relative velocity.

[0028] Step 3: Calculate the target's maximum lateral relative velocity. Set a reasonable threshold for the maximum lateral absolute velocity of the target. (For example, based on the maximum reasonable speed of movement of a pedestrian or cyclist), calculate the target's maximum lateral relative velocity:

[0029] If the camera can acquire the orientation information of the target, the maximum lateral relative velocity of the target is calculated using the following formula:

[0030] If the camera cannot obtain the target's orientation information, the target's maximum lateral relative velocity is calculated using the following formula:

[0031] in, This represents the target's maximum lateral relative velocity. This represents the preset threshold for the maximum absolute lateral velocity of the target. This represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system.

[0032] This step considers the target's maximum possible relative lateral velocity under kinematic constraints, with the aim of providing the necessary input for the next step of calculating the maximum lateral distance change between frames.

[0033] Step 4: Calculate the maximum lateral distance change between frames. According to step three Given the sampling frequency v, calculate the theoretically maximum allowable change in the lateral distance of the target between two adjacent frames. :

[0034] in, This represents the maximum change in the lateral distance between targets between frames. denoted by , where v represents the target's maximum lateral relative velocity, and v represents the sampling frequency of the target information.

[0035] In this step, the principle of determining whether the lateral distance has changed is to determine whether the difference in lateral distance between two frames output by the camera (i.e., the change in the lateral distance of the target between frames) is greater than the maximum change in the lateral distance of the target between frames calculated in step four. This embodiment considers that under normal circumstances, the change in the lateral distance of the target between frames should not be greater than the maximum possible change in the lateral distance of the target between frames under kinematic constraints.

[0036] Step 5: Jump Detection and Flag Output The target lateral distance between frames is obtained by comparing it with the maximum change in the target lateral distance between frames, and the target lateral information confidence flag function f(x) is obtained. ):

[0037] If f=0, it means the horizontal information is unreliable; if f=1, it means the horizontal information is reliable. in, Indicates the horizontal distance of the target in the current frame. This indicates the horizontal distance of the target in the previous frame. This indicates the maximum change in the horizontal distance between targets in different frames.

[0038] The collision risk assessment module of the AEB system can use this flag f to decide whether to adopt the current lateral motion state of the target for risk assessment. When f=0, the current lateral information can be ignored or used cautiously, thereby avoiding false braking caused by abrupt changes in perception.

[0039] In this embodiment, since the camera outputs the target's lateral distance at a certain frame rate, if the inter-frame target lateral distance output by the camera is greater than the maximum possible value under kinematic constraints, it is considered that the output has a jump and is unreliable. In order to obtain the maximum possible value of the inter-frame change in the target's lateral distance under kinematic constraints and thus complete the judgment of the reliability of the camera's lateral information, firstly, the target's lateral dragging velocity (i.e., the projection of the target's dragging velocity on the X-axis) is calculated. Then, a maximum absolute lateral motion velocity of the target is set, and the maximum relative lateral velocity of the target is calculated based on the relationship between the absolute velocity, dragging velocity, and relative velocity. Finally, the maximum possible value of the inter-frame change in the target's lateral distance can be calculated based on the target's maximum relative lateral velocity and the camera's output frame rate, thereby obtaining the target's lateral information confidence flag function and realizing the accurate judgment of the jump in the lateral distance when crossing the target.

[0040] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining a jump in lateral distance when crossing a target, characterized in that, include: Target information is acquired through a data acquisition device, and the target information includes the target's lateral distance, longitudinal distance, and azimuth angle in the vehicle coordinate system. The yaw rate of the vehicle is obtained, and based on the yaw rate of the vehicle, the lateral distance, longitudinal distance, and azimuth angle, the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system is calculated. The maximum lateral relative velocity of the target is calculated based on the preset threshold of the absolute velocity of the target's maximum lateral movement and the projected velocity. Calculate the maximum change in the target's lateral distance between frames based on the target's maximum relative lateral velocity and the sampling frequency of the target information; Calculate the change in the target's lateral distance between the current frame and the previous frame, compare the change in the target's lateral distance with the maximum change in the target's lateral distance between frames, and output the target's lateral information confidence flag based on the comparison result.

2. The method for determining lateral distance jumps across a target according to claim 1, characterized in that, The projected velocity of the target's entrainment velocity in the X direction of the vehicle coordinate system is calculated by the following formula: in, ω represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system, ω represents the yaw rate of the vehicle, x represents the lateral distance of the target in the vehicle's coordinate system, y represents the longitudinal distance of the target in the vehicle's coordinate system, and θ represents the azimuth angle of the target in the vehicle's coordinate system.

3. The method for determining lateral distance jumps across a target according to claim 1, characterized in that, If the acquisition device can obtain the orientation information of the target, the maximum lateral relative velocity of the target is calculated using the following formula: in, This represents the target's maximum lateral relative velocity. This represents the preset threshold for the maximum absolute lateral velocity of the target. This represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system.

4. The method for determining lateral distance jumps across a target according to claim 1, characterized in that, If the acquisition device cannot obtain the orientation information of the target, the maximum lateral relative velocity of the target is calculated using the following formula: in, This represents the target's maximum lateral relative velocity. This represents the preset threshold for the maximum absolute lateral velocity of the target. This represents the projected velocity of the target's entrainment velocity in the X direction of the vehicle's coordinate system.

5. The method for determining lateral distance jumps across a target according to claim 1, characterized in that, The maximum change in the inter-frame target lateral distance is calculated using the following formula: in, This represents the maximum change in the lateral distance between targets between frames. denoted by , where v represents the target's maximum lateral relative velocity, and v represents the sampling frequency of the target information.

6. The method for determining lateral distance jumps across a target according to claim 1, characterized in that, The function f(x, of the target lateral information confidence flag bit) The definition is: if |x(k) - x(k-1)| > 1 / 2 If f=0, it means the horizontal information is unreliable; if |x(k) - x(k-1)|≤ If f=1, it indicates that the horizontal information is reliable; in, Indicates the horizontal distance of the target in the current frame. This indicates the horizontal distance of the target in the previous frame. This indicates the maximum change in the horizontal distance between targets in different frames.

7. The method for determining lateral distance jumps across a target according to claim 1, characterized in that, The yaw rate of the vehicle is obtained through the CAN bus of the vehicle body.

Images