Automatic driving vehicle obstacle avoidance method

By generating different number of collision detection circles according to vehicle speed conditions and using ray method to determine whether obstacles and vehicles collide, the accuracy and timeliness of collision detection algorithms in the prior art under limited system resources are solved, and more efficient collision detection warning.

CN118938894APending Publication Date: 2024-11-12FORYOU GENERAL ELECTRONICS
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410803155.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-20
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing collision detection algorithms are difficult to achieve accurate and timely collision warning when the system resources are limited. The external circular method excessively limits the travelable area, AABB ignores the rotation of the object, and OBB has high computational complexity, resulting in large system resource consumption.

Method used

By generating different number of collision detection circles according to vehicle speed conditions, matching different obstacle collision detection strategies, using the ray method to determine whether the obstacles collide with the vehicle, improving the accuracy and timeliness of collision detection.

Benefits of technology

Under the conditions of limited system resources, more accurate and timely collision detection and early warning are achieved, system resources are fully utilized, and the collision detection capabilities of intelligent driving systems are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938894A_ABST
    Figure CN118938894A_ABST
Patent Text Reader

Abstract

The invention provides an automatic driving vehicle obstacle avoidance method. The method comprises the following steps: step 1, acquiring a target driving path of a vehicle; step 2, acquiring a planned vehicle speed of the vehicle at any point in a preset range on the target path and a corresponding collision detection circle; step 3, acquiring coordinates of all obstacles in the preset range; step 4, calculating linear distances from the center of the collision detection circle to all obstacles; 5, whether the linear distance is larger than the radius of the collision detection circle or not is judged, if yes, it is judged that the vehicle has no collision risk, and if not, it is judged that the vehicle has the collision risk; and 6, judging whether the vehicle can generate a new target driving path or not, if so, returning to the step 2, and otherwise, sending preset prompt information. According to the invention, the intelligent driving system can improve the accuracy and timeliness of collision detection early warning under the condition of limited system resources.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to an obstacle avoidance method for an autonomous driving vehicle. Background Art

[0002] With the development of technology, smart cars with autonomous driving capabilities are attracting more and more attention from car companies and consumers.

[0003] Because smart cars use technologies such as sensors, computer vision, and artificial intelligence, they can achieve safer and more precise driving and navigation, and better adapt to changes in road conditions and road traffic pressure. Therefore, smart cars can greatly improve human travel experience and efficiency.

[0004] Among the many technologies for realizing intelligent driving, obstacle avoidance algorithms are widely used as important algorithms in the fields of unmanned driving, robotics, etc. The main purpose is to enable robots or unmanned vehicles to navigate autonomously and avoid obstacles. An important technical point for realizing obstacle avoidance is the collision detection algorithm. The current collision detection algorithms include the circumscribed circle method, AABB (axis-aligned bounding box), SAT / OBB (separating axis theorem / directional bounding box), etc. Although the circumscribed circle method is intuitive, it may overly restrict the drivable area; although AABB is simple to calculate, it ignores the rotation of the object and is therefore not accurate enough; although OBB takes into account the rotation of the object and provides more accurate collision detection, the computational complexity also increases accordingly, and the system resource consumption is very large. Summary of the invention

[0005] The present invention provides an obstacle avoidance method for an autonomous driving vehicle, aiming to solve the defects in the prior art and enable an intelligent driving system to improve the accuracy and timeliness of collision detection and warning under the condition of limited system resources.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] The present invention provides an obstacle avoidance method for an autonomous driving vehicle, comprising:

[0008] Step 1: Obtain the target driving path of the vehicle;

[0009] Step 2: Obtain the planned speed of the vehicle at any point within a preset range on the target path and the corresponding collision detection circle;

[0010] Step 3: Obtain the coordinates of all obstacles within the preset range;

[0011] Step 4: Calculate the straight-line distance from the center of all the collision detection circles to all obstacles;

[0012] Step 5: determine whether the straight-line distance is greater than the radius of the collision detection circle. If so, determine that the vehicle has no collision risk; otherwise, determine that the vehicle has a collision risk.

[0013] Step 6: Determine whether the vehicle can generate a new target driving path. If yes, return to step 2; otherwise, issue a preset prompt message.

[0014] Specifically, the method for determining the collision detection circle includes:

[0015] Step 201, obtaining target parameters of the vehicle, and establishing a vehicle rectangular model according to the target parameters, wherein the target parameters include vehicle length, width, and vehicle rear overhang length;

[0016] Step 202, establishing a global coordinate system, wherein the global coordinate system takes the center position of the rear axle at the time of vehicle startup as the origin, the Y-axis pointing to the front of the vehicle as the positive direction, and the X-axis pointing to the right side of the vehicle as the positive direction;

[0017] Step 203, obtaining the position and posture of the vehicle in the global coordinate system;

[0018] Step 204: Obtain the planned speed of the vehicle at any point within a preset range on the target path, and determine the corresponding collision detection circle according to the planned speed in accordance with preset rules.

[0019] Specifically, the preset rule is that the vehicle speed is negatively correlated with the number of collision detection circles.

[0020] Specifically, the step 204 includes:

[0021] A) when the vehicle speed is greater than a first preset threshold, determining the collision detection circle according to a first rule;

[0022] B) when the vehicle speed is less than a first preset threshold but greater than a second preset threshold, determining the collision detection circle according to a second rule;

[0023] C) when the vehicle speed is less than the second preset threshold but greater than the third preset threshold, determining the collision detection circle according to a third rule;

[0024] D) when the vehicle speed is less than the third preset threshold but greater than the fourth preset threshold, determining the collision detection circle according to a fourth rule.

[0025] Specifically, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold decrease in sequence.

[0026] Specifically, the first rule is: the collision detection circle is a first collision detection circle, and the radius and center coordinates of the collision detection circle are determined according to a first preset formula, and the first preset formula is:

[0027]

[0028] Among them, R 1 is the radius of the first collision detection circle, (x o1 ,y o1 ) represents the center coordinates of the first collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0029] Specifically, the second rule is: the collision detection circle includes a second collision detection circle and a third collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a second preset formula, and the second preset formula is:

[0030]

[0031] Among them, R 2 is the radius of the second collision detection circle and the third collision detection circle, (x o21 ,y o21 ) represents the center coordinates of the second collision detection circle, (x o22 ,y o22 ) represents the center coordinates of the third collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0032] Specifically, the third rule is: the collision detection circle includes a fourth collision detection circle, a fifth collision detection circle, and a sixth collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a third preset formula, and the third preset formula is:

[0033]

[0034] Among them, R 3 is the radius of the fourth collision detection circle, the fifth collision detection circle, and the sixth collision detection circle, (x o31 ,y o31 ) represents the center coordinates of the fourth collision detection circle, (x o32 ,y o32 ) represents the center coordinates of the fifth collision detection circle, (x o33 ,y o33 ) represents the center coordinates of the sixth collision detection circle, (x r ,y r) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0035] Specifically, the fourth rule is: the collision detection circle includes a seventh collision detection circle, an eighth collision detection circle, a ninth collision detection circle, and a tenth collision detection circle with equal radii, and the radius and the center coordinates of the collision detection circle are determined according to a fourth preset formula, and the fourth preset formula is:

[0036]

[0037] Among them, R 4 is the radius of the seventh collision detection circle, the eighth collision detection circle, the ninth collision detection circle, and the tenth collision detection circle, (x o41 ,y o41 ) represents the center coordinates of the seventh collision detection circle, (x o42 ,y o42 ) represents the center coordinates of the eighth collision detection circle, (x o43 ,y o43 ) represents the center coordinates of the ninth collision detection circle, (x o44 ,y o44 ) represents the center coordinates of the tenth collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0038] Furthermore, after step 1 and before step 2, the method further includes:

[0039] Step 10: Determine whether the current vehicle speed is lower than a fourth preset threshold or is in a reversing state. If so, determine whether an obstacle collides with the vehicle based on the ray method. Otherwise, proceed to the next step.

[0040] The beneficial effect of the present invention is that the present invention generates a corresponding number of collision detection circles according to different vehicle speed conditions, thereby matching different obstacle collision detection strategies, which can not only achieve accurate detection but also make full use of system resources, so that the intelligent driving system can improve the accuracy and timeliness of collision detection warning under the condition of limited system resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of an obstacle avoidance method for an autonomous driving vehicle of the present invention;

[0042] Figure 2A is a schematic diagram of determining a collision detection circle according to a first rule of the present invention;

[0043] Figure 2Bis a schematic diagram of determining a collision detection circle according to a second rule of the present invention;

[0044] Figure 2C is a schematic diagram of determining a collision detection circle according to the third rule of the present invention;

[0045] Figure 2D It is a schematic diagram of determining the collision detection circle according to the fourth rule of the present invention. DETAILED DESCRIPTION

[0046] The following specifically describes the embodiments of the present invention in conjunction with the accompanying drawings, which are only for reference and illustration and do not constitute a limitation on the scope of patent protection of the present invention.

[0047] In the process described in the specification, claims or drawings of the present invention, the serial numbers of each step (such as step 10, 20, etc.) are included. The serial numbers are only used to distinguish the steps, and the serial numbers themselves do not represent any execution order. It should be noted that the descriptions such as "first" and "second" in this article are only used to distinguish the description objects, etc., and do not represent the order of precedence, nor do they mean that "first", "second", etc. are different types.

[0048] Example 1

[0049] As shown in the figure, this embodiment provides an obstacle avoidance method for an autonomous driving vehicle, including:

[0050] Step 1: Obtain the target driving path of the vehicle.

[0051] If the vehicle is in the automatic driving state, the automatic driving system will plan and generate multiple possible vehicle driving paths, and then select the best one as the target driving path through the vehicle's kinematic model, and then the vehicle control system will control the vehicle to track the target driving path.

[0052] Step 2: Obtain the planned speed of the vehicle at any point within a preset range on the target path and the corresponding collision detection circle.

[0053] In specific implementation, the preset range can be set according to actual conditions, for example, 20 meters to 50 meters.

[0054] In this embodiment, the method for determining the collision detection circle includes:

[0055] Step 201: Obtain target parameters of the vehicle and establish a rectangular model of the vehicle according to the target parameters. The target parameters include vehicle length L, width W, vehicle rear overhang length L r .

[0056] Step 202, establish a global coordinate system, wherein the global coordinate system takes the center position of the rear axle at the moment of vehicle startup as the origin, the Y-axis pointing to the front of the vehicle as the positive direction, and the X-axis pointing to the right side of the vehicle as the positive direction.

[0057] Step 203: Obtain the position and posture of the vehicle in the global coordinate system.

[0058] As shown in FIG2 , the vehicle's position and posture are expressed as (x, y, θ), where x and y represent the coordinates of the midpoint of the rear axle of the vehicle body, and θ represents the angle between the front of the vehicle and the x-axis of the coordinate system, that is, the yaw angle.

[0059] Step 204: Obtain the planned speed of the vehicle at any point within a preset range on the target path, and determine the corresponding collision detection circle according to the planned speed in accordance with preset rules.

[0060] In this embodiment, the preset rule is that the vehicle speed is negatively correlated with the number of collision detection circles.

[0061] In this embodiment, step 204 includes:

[0062] A) when the vehicle speed is greater than a first preset threshold, determining the collision detection circle according to a first rule;

[0063] B) when the vehicle speed is less than a first preset threshold but greater than a second preset threshold, determining the collision detection circle according to a second rule;

[0064] C) when the vehicle speed is less than the second preset threshold but greater than the third preset threshold, determining the collision detection circle according to a third rule;

[0065] D) when the vehicle speed is less than the third preset threshold but greater than the fourth preset threshold, determining the collision detection circle according to a fourth rule.

[0066] In this embodiment, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold are sequentially decreased.

[0067] As an implementable example, the first preset threshold is 80 km / h, the second preset threshold is 60 km / h, the third preset threshold is 40 km / h, and the fourth preset threshold is 20 km / h.

[0068] like Figure 2A As shown, in this embodiment, the first rule is: the collision detection circle is a first collision detection circle, and the radius and center coordinates of the collision detection circle are determined according to a first preset formula, and the first preset formula is:

[0069]

[0070] Among them, R 1is the radius of the first collision detection circle, (x o1 ,y o1 ) represents the center coordinates of the first collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0071] like Figure 2B As shown, in this embodiment, the second rule is: the collision detection circle includes a second collision detection circle and a third collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a second preset formula, and the second preset formula is:

[0072]

[0073] Among them, R 2 is the radius of the second collision detection circle and the third collision detection circle, (x o21 ,y o21 ) represents the center coordinates of the second collision detection circle, (x o22 ,y o22 ) represents the center coordinates of the third collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0074] like Figure 2C As shown, in this embodiment, the third rule is: the collision detection circle includes a fourth collision detection circle, a fifth collision detection circle, and a sixth collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a third preset formula, and the third preset formula is:

[0075]

[0076] Among them, R 3 is the radius of the fourth collision detection circle, the fifth collision detection circle, and the sixth collision detection circle, (x o31 ,y o31 ) represents the center coordinates of the fourth collision detection circle, (x o32 ,y o32 ) represents the center coordinates of the fifth collision detection circle, (x o33 ,y o33 ) represents the center coordinates of the sixth collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0077] like Figure 2D As shown, in this embodiment, the fourth rule is: the collision detection circle includes a seventh collision detection circle, an eighth collision detection circle, a ninth collision detection circle, and a tenth collision detection circle with equal radii, and the radius and the center coordinates of the collision detection circle are determined according to a fourth preset formula, and the fourth preset formula is:

[0078]

[0079] Among them, R 4 is the radius of the seventh collision detection circle, the eighth collision detection circle, the ninth collision detection circle, and the tenth collision detection circle, (x o41 ,y o41 ) represents the center coordinates of the seventh collision detection circle, (x o42 ,y o42 ) represents the center coordinates of the eighth collision detection circle, (x o43 ,y o43 ) represents the center coordinates of the ninth collision detection circle, (x o44 ,y o44 ) represents the center coordinates of the tenth collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

[0080] Step 3: Obtain the coordinates P of all obstacles within the preset range j .

[0081] Step 4: Calculate the center O of all the collision detection circles i Straight-line distance D to all obstacles ij .

[0082] Step 5: Determine the straight-line distance D ij Is it greater than the radius R of the collision detection circle? i If yes, the vehicle is judged to have no collision risk; otherwise, the vehicle is judged to have collision risk.

[0083] Step 6: Determine whether the vehicle can generate a new target driving path. If yes, return to step 2; otherwise, issue a preset prompt message.

[0084] In a specific implementation, the preset information may be information to remind the driver to take over the vehicle so that the driver can take corresponding actions.

[0085] Example 2

[0086] Different from the first embodiment, the present embodiment further includes, after the step 1:

[0087] Step 10: Determine whether the current vehicle speed is lower than a fourth preset threshold or is in a reversing state. If so, determine whether an obstacle collides with the vehicle based on the ray method. Otherwise, proceed to the next step.

[0088] Judging whether an obstacle collides with a vehicle based on the ray method is an existing technology. Its basic principle is to determine the straight line equations of the four borders of the vehicle based on the coordinates of the four vertices of the vehicle model, so as to obtain the vertex of the ray with the obstacle, derive rays of arbitrary slopes, and then solve the coordinates of the intersection of the ray and the straight line equations of the four borders of the vehicle. Finally, it is judged whether the obstacle to be judged is inside the vehicle model based on the number of intersections.

[0089] The above disclosure is only a preferred embodiment of the present invention, which cannot be used to limit the scope of protection of the present invention. Therefore, equivalent changes made according to the scope of the patent application of the present invention are still within the scope covered by the present invention.

Claims

1. An obstacle avoidance method for an autonomous driving vehicle, It is characterized in that include: Step 1: Obtain the target driving path of the vehicle; Step 2: Obtain the planned speed of the vehicle at any point within a preset range on the target path and the corresponding collision detection circle; Step 3: Obtain the coordinates of all obstacles within the preset range; Step 4: Calculate the straight-line distance from the center of all the collision detection circles to all obstacles; Step 5: determine whether the straight-line distance is greater than the radius of the collision detection circle. If so, determine that the vehicle has no collision risk; otherwise, determine that the vehicle has a collision risk. Step 6: Determine whether the vehicle can generate a new target driving path. If yes, return to step 2; otherwise, issue a preset prompt message.

2. The method for avoiding obstacles of an autonomous driving vehicle according to claim 1, It is characterized in that The method for determining the collision detection circle includes: Step 201, obtaining target parameters of the vehicle, and establishing a vehicle rectangular model according to the target parameters, wherein the target parameters include vehicle length, width, and vehicle rear overhang length; Step 202, establishing a global coordinate system, wherein the global coordinate system takes the center position of the rear axle at the time of vehicle startup as the origin, the Y-axis pointing to the front of the vehicle as the positive direction, and the X-axis pointing to the right side of the vehicle as the positive direction; Step 203, obtaining the position and posture of the vehicle in the global coordinate system; Step 204: Obtain the planned speed of the vehicle at any point within a preset range on the target path, and determine a corresponding collision detection circle according to the planned speed in accordance with preset rules.

3. The automatic driving vehicle obstacle avoidance method according to claim 2, It is characterized in that The preset rule is that the vehicle speed is negatively correlated with the number of collision detection circles.

4. The automatic driving vehicle obstacle avoidance method according to claim 3, It is characterized in that The step 204 includes: A) when the vehicle speed is greater than a first preset threshold, determining the collision detection circle according to a first rule; B) when the vehicle speed is less than a first preset threshold but greater than a second preset threshold, determining the collision detection circle according to a second rule; C) when the vehicle speed is less than the second preset threshold but greater than the third preset threshold, determining the collision detection circle according to a third rule; D) when the vehicle speed is less than the third preset threshold but greater than the fourth preset threshold, determining the collision detection circle according to a fourth rule.

5. The method for avoiding obstacles of an autonomous driving vehicle according to claim 4, It is characterized in that The first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold decrease in sequence.

6. The automatic driving vehicle obstacle avoidance method according to claim 5, It is characterized in that The first rule is: the collision detection circle is a first collision detection circle, and the radius and center coordinates of the collision detection circle are determined according to a first preset formula, and the first preset formula is: Among them, R 1 is the radius of the first collision detection circle, (x o1 ,y o1 ) represents the center coordinates of the first collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

7. The method for avoiding obstacles of an autonomous driving vehicle according to claim 5, It is characterized in that The second rule is: the collision detection circle includes a second collision detection circle and a third collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a second preset formula, and the second preset formula is: Among them, R 2 is the radius of the second collision detection circle and the third collision detection circle, (x o21 ,y o21 ) represents the center coordinates of the second collision detection circle, (x o22 ,y o22 ) represents the center coordinates of the third collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

8. The method for avoiding obstacles of an autonomous driving vehicle according to claim 5, It is characterized in that The third rule is: the collision detection circle includes a fourth collision detection circle, a fifth collision detection circle, and a sixth collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a third preset formula, and the third preset formula is: Among them, R 3 is the radius of the fourth collision detection circle, the fifth collision detection circle, and the sixth collision detection circle, (x o31 ,y o31 ) represents the center coordinates of the fourth collision detection circle, (x o32 ,y o32 ) represents the center coordinates of the fifth collision detection circle, (x o33 ,y o33 ) represents the center coordinates of the sixth collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

9. The method for avoiding obstacles of an autonomous driving vehicle according to claim 5, It is characterized in that The fourth rule is: the collision detection circle includes a seventh collision detection circle, an eighth collision detection circle, a ninth collision detection circle, and a tenth collision detection circle with equal radii, and the radius and center coordinates of the collision detection circle are determined according to a fourth preset formula, and the fourth preset formula is: Among them, R 4 is the radius of the seventh collision detection circle, the eighth collision detection circle, the ninth collision detection circle, and the tenth collision detection circle, (x o41 ,y o41 ) represents the center coordinates of the seventh collision detection circle, (x o42 ,y o42 ) represents the center coordinates of the eighth collision detection circle, (x o43 ,y o43 ) represents the center coordinates of the ninth collision detection circle, (x o44 ,y o44 ) represents the center coordinates of the tenth collision detection circle, (x r ,y r ) represents the coordinate of the midpoint of the rear edge of the vehicle body, L represents the length of the vehicle, W represents the width of the vehicle, and θ represents the yaw angle of the vehicle.

10. The automatic driving vehicle obstacle avoidance method according to claim 1, It is characterized in that After step 1 and before step 2, the method further includes: Step 10: Determine whether the current vehicle speed is lower than a fourth preset threshold or is in a reversing state. If so, determine whether an obstacle collides with the vehicle based on the ray method. Otherwise, proceed to the next step.

Citation Information

Cited By

  • Method and system for preventing rear-end collision of automobile

    CN119611354A