Parking control method based on path tracking and terminal equipment

Abstract

The invention relates to the technical field of intelligent auxiliary driving, and provides a parking control method based on path tracking and terminal equipment, during parking control, a pre-planned parking planning path is taken as a standard, a tracking path point is screened out from the path through a vehicle pose, and actuator oscillation caused by sudden change of a steering angle is avoided; further, an instantaneous tracking steering circular curve is calculated according to the tracking path points and the adjacent preorder path points and postorder path points, preview points are determined from the tracking steering circular curve based on the preview distance, and by matching the curvature of the path points, on one hand, the preview points meeting kinematics constraints and the path curvature adapting to the moment change are screened out, and on the other hand, the path curvature adapting to the moment change is screened out; on the other hand, tracking errors are effectively restrained, the tracking calculation precision is improved, then the parking efficiency in a compact parking space scene is improved, the parking experience is improved, and driving habits of a driver are better met.

Description

Technical Field

[0001] This invention relates to the field of intelligent assisted driving technology, and in particular to a parking control method and terminal device based on path tracking. Background Technology

[0002] Currently, commonly used automatic parking systems mainly consist of the following three parts: parking space recognition system, path planning system, and parking control system.

[0003] In parking control systems, vehicle path tracking control refers to the process by which a vehicle, under the combined action of lateral and longitudinal motion controllers, approaches a reference path. The specific techniques of vehicle path tracking control vary depending on the application scenario. Common path tracking control methods used in autonomous parking systems include PID control algorithms, PurePursuit control algorithms, model predictive control algorithms, and Stanley control algorithms.

[0004] Currently, there are disadvantageous parking scenarios where parking spaces are small: if the parking space is small, the vehicle may collide with other vehicles or obstacles, thus affecting the success rate of autonomous parking.

[0005] In intelligent parking systems, once a parking path is planned, the vehicle needs to be controlled to travel along the desired path; this is the job of the path tracking system. Therefore, the performance of the path tracking algorithm directly determines the actual driving path of the vehicle, and thus directly determines the success or failure of intelligent parking. However, existing path tracking algorithms do not consider the response delay of the vehicle's steering mechanism. Therefore, in situations with limited parking space, or where the path curves or curvature changes significantly, the path tracking performance will be poor. Summary of the Invention

[0006] This invention provides a parking control method and terminal device based on path tracking, which solves the technical problems of existing path tracking algorithms having response delays, being unable to adapt to parking scenarios with limited parking space, having poor path tracking performance, and failing to control parking.

[0007] To address the above technical problems, this invention provides a parking control method based on path tracing, comprising: A vehicle kinematics model is established, and a parking planning path is obtained; wherein, the parking planning path contains multiple discrete parking path points; The vehicle pose is determined based on the vehicle kinematics model, and tracking path points are selected from the parking path points of the parking planning path based on the vehicle pose. Obtain the preceding and following path points adjacent to the tracking path point, and calculate the instantaneous tracking turning circular curve based on the preceding path point, the tracking path point, and the following path point; The aiming point is determined from the tracking turning circular curve based on the aiming distance; The target steering wheel angle is calculated based on the pre-aiming point for vehicle control.

[0008] This basic solution, when performing parking control, uses a pre-planned parking path as the standard. It selects tracking path points from the path based on the vehicle's pose, avoiding actuator oscillations caused by sudden changes in steering angle. Then, it calculates the instantaneous tracking steering curve based on the tracking path points and adjacent preceding and following path points. Based on the pre-aiming distance, it determines the pre-aiming point from the tracking steering curve. By matching the curvature of the path points, it selects pre-aiming points that meet kinematic constraints, adapts to the constantly changing path curvature, and effectively suppresses tracking errors, improving tracking calculation accuracy. This, in turn, improves parking efficiency in compact parking scenarios, enhances the parking experience, and better suits the driver's driving habits.

[0009] In a further implementation, the vehicle pose is determined based on the vehicle kinematic model, and tracking path points are selected from the parking path points of the parking planning path based on the vehicle pose, including: Obtain the path segment closest to the vehicle's current position from the parking planning path, and obtain all the parking path points on the path segment; Based on the vehicle kinematics model, the current vehicle pose is determined; Based on the vehicle kinematics model and the vehicle's current position, calculate the vehicle's pose orientation vector relative to each parking path point; Based on the pose orientation vector and the current vehicle pose, a tracking path point that is consistent with the current path direction of the vehicle is selected from the parking path points.

[0010] In a further implementation, based on the pose orientation vector and the current vehicle pose, tracking path points that are consistent with the current path direction of the vehicle are selected from the parking path points, specifically: Iterate through the current vehicle pose at each parking path point and calculate the inner product of the current vehicle pose and the pose orientation vector; iterate through the inner product of each parking path point, and when the first inner product greater than zero is detected, determine that the parking path point corresponding to this inner product is consistent with the current path direction of the vehicle, and determine the parking path point as the tracking path point.

[0011] When selecting tracking path points, this scheme calculates the vehicle's pose orientation vector relative to each parking path point, calculates the dot product of the current vehicle pose and the pose orientation vector, and obtains the first parking path point with a dot product greater than zero. This is used as the tracking path point selected from the parking path points that is consistent with the vehicle's current path direction. The difference between the parking path point and the vehicle's driving direction is identified through vector dot product calculation, and reverse / unreachable path points are excluded. The driving direction switching point is automatically identified to ensure that the vehicle can be smoothly tracked at the selected starting point. While ensuring real-time performance, the directional robustness of the control system is significantly improved.

[0012] In a further implementation, based on the vehicle kinematics model and the vehicle's current position, the pose orientation vector of the vehicle relative to each parking path point is calculated. Specifically, based on the vehicle kinematics model, the position coordinates of the vehicle's current position and the parking coordinates of each parking path point are determined, and the pose orientation vector of the vehicle relative to each parking path point is calculated with the position coordinates as the starting point and the parking coordinates as the ending point.

[0013] This solution calculates the vehicle's pose orientation vector relative to each parking path point based on the parking coordinates of each parking path point, with the position coordinates as the starting point and the parking coordinates as the ending point. This provides the adjustment and adaptation degree of the parking path point relative to the vehicle. The positive and negative values ​​of the vectors are used to quantify the data, resulting in low computational difficulty and high computational efficiency.

[0014] In a further implementation, obtaining the preceding and following path points adjacent to the tracking path point, and calculating the instantaneous tracking turning circular curve based on the preceding path point, the tracking path point, and the following path point, includes: Based on the parking path point corresponding to the tracking path point, the previous parking path point adjacent to the tracking path point is obtained from the parking planning path as the preceding path point, and the next parking path point adjacent to the tracking path point is obtained from the parking planning path as the following path point. Substitute the preceding path point, the tracking path point, and the following path point into the general equation of a circle to obtain the instantaneous tracking turning circle curve, and determine the center coordinates and the turning circle radius.

[0015] In a further embodiment, determining the aiming point from the tracking turning circular curve based on the aiming distance includes: Determine the position of the rear axle center point based on the vehicle body parameters; Extract a parking planning curve from the instantaneous tracking steering circular curve that lies between the preceding path point and the following path point; Draw a circle with the center point of the rear axle of the vehicle as the starting point and the aiming distance as the radius, and obtain the intersection point of the circle with the parking planning curve as the aiming point.

[0016] When selecting the aiming point, this scheme first draws a tracking steering circle curve based on the parking path points (previous path points, tracking path points, and subsequent path points). Then, it draws a circle with the rear axle center point of the vehicle as the starting point and the aiming distance as the radius, and obtains the intersection point with the parking planning curve as the aiming point. On the one hand, it uses the parking path points as a basis to realize the tracking control of the parking planning path as much as possible. On the other hand, it dynamically selects the aiming point based on the aiming distance, which fully fits the current vehicle posture (avoiding multiple adjustments), improves the aiming accuracy, reduces the difficulty of tracking steering, and further realizes path smoothing optimization.

[0017] In a further implementation, the aiming distance is directly proportional to the vehicle's speed, and the aiming distance is determined in real time based on the vehicle's current speed.

[0018] This solution calibrates the pre-aiming distance to be proportional to the vehicle's speed. At high speeds, the pre-aiming distance is increased to compensate for control delay, while at low speeds, the pre-aiming distance is shortened to improve accuracy. This achieves simultaneous improvement in delay compensation and stability, effectively increasing parking efficiency.

[0019] In a further implementation scheme, the target steering wheel angle is calculated based on the pre-aiming point, specifically as follows: Based on the vehicle kinematics model, the current vehicle pose is determined, including the current heading angle and position coordinates; Obtain the pre-aiming coordinates of the pre-aiming point; The target steering wheel angle is calculated based on the current heading angle, position coordinates, aiming coordinates, and aiming distance, and vehicle control is performed based on the target steering wheel angle.

[0020] In a further implementation scheme, the formula for calculating the target steering wheel angle is as follows:

[0021] In the formula, Indicates the target steering wheel angle. This indicates the aiming distance, and L represents the vehicle's wheelbase. Indicates the aiming distance The angle between the direction and the axle.

[0022] This solution calculates the target steering wheel angle based on the current heading angle, position coordinates, aiming coordinates, and aiming distance, and controls the vehicle based on the target steering wheel angle, strictly following the bicycle kinematics model and effectively avoiding kinematic distortion. The present invention also provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a parking control method based on path tracing as described above. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the parking control method based on path tracing provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the vehicle kinematics model provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the parking planning path provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the pose orientation vector provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the instantaneous tracking steering circular curve provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the pre-aiming point selection based on the instantaneous tracking steering circular curve provided in an embodiment of the present invention. Detailed Implementation

[0024] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0025] Example 1 This invention provides a parking control method based on path tracing, such as... Figure 1 As shown, in this embodiment, it includes: S1. Establish a vehicle kinematics model and obtain a parking planning path; wherein, the parking planning path contains multiple discrete parking path points; See Figure 2 The specific steps for establishing the vehicle kinematic model are as follows: A right-handed coordinate system is established with the vehicle's position at a certain moment of motion (e.g., the moment of power-on) as the origin and the vehicle's initial direction of travel as the positive x-axis. Figure 2 This is a schematic diagram of the vehicle in motion. The angle of the vehicle body is the angle between the direction of the front of the vehicle and the positive x-axis.

[0026] Assuming the upstream path planning module has already obtained the vehicle's parking plan path based on the actual vehicle and environmental conditions, as shown in Figure 3, it is represented as a series of discrete point combinations. These discrete points can be marked as needed. .

[0027] S2. Determine the vehicle pose based on the vehicle kinematic model, and select tracking path points from the parking path points of the parking planning path based on the vehicle pose, including: S21. Obtain the path segment closest to the current position of the vehicle from the parking planning path, and obtain all the parking path points on the path segment; S22. Determine the current vehicle pose based on the vehicle kinematics model; Based on the steering ratio k between the steering wheel and the front wheels, the input steering wheel angle... (Positive values ​​to the left, negative values ​​to the right), calculate the angle of the front wheel:

[0028] Let R be the radius of rotation of the rear axle center of the vehicle, and let the yaw angle of the vehicle be... The vehicle's speed is .

[0029] Based on Ackermann kinematics, a kinematic model of the tractor is established, assuming the vehicle's pose at time k-1 is... Then at time k (sampling time interval is...) The pose of the ) is as follows:

[0030] The current position of the vehicle can be obtained from the above, denoted as . .

[0031] S23. Based on the vehicle kinematics model and the vehicle's current position, calculate the vehicle's pose orientation vector relative to each parking path point, specifically: Based on the vehicle kinematics model, the current position coordinates of the vehicle and the parking coordinates of each parking path point are determined. The pose orientation vector of the vehicle relative to each parking path point is calculated with the position coordinates as the starting point and the parking coordinates as the ending point.

[0032] This embodiment calculates the vehicle's pose orientation vector relative to each parking path point based on the parking coordinates of each parking path point, with the position coordinates as the starting point and the parking coordinates as the ending point. This provides the adjustment and adaptation degree of the parking path point relative to the vehicle. The positive and negative values ​​of the vector are used to quantify the data, which is easy to calculate and has high computational efficiency.

[0033] S24. Based on the pose orientation vector and the current vehicle pose, select tracking path points from the parking path points that are consistent with the current path direction of the vehicle, specifically: Iterate through the current vehicle pose at each parking path point and calculate the inner product of the current vehicle pose and the pose orientation vector; iterate through the inner product of each parking path point, and when the first inner product greater than zero is detected, determine that the parking path point corresponding to this inner product is consistent with the current path direction of the vehicle, and determine the parking path point as the tracking path point.

[0034] Specifically: See Figure 4 At a certain moment, the vehicle obtains the parking coordinates of the parking path points and the position coordinates of the vehicle's current position. Then, a vector is constructed with the vehicle's current position as the starting point and all parking path points (discrete points on the path) as the ending points.

[0035] Hollow dots in the image The black arrows represent the vehicle's heading angle, and the solid dots together indicate the vehicle's current position. . This is the vector representation of the vehicle's pose in the coordinate system.

[0036] Starting from the first parking path point, calculate the vector formed by the vehicle's current position and a given parking path point, and the vehicle's orientation vector. The inner product of the two paths is used to determine the next tracking path point for the vehicle when the first inner product is greater than zero. Assuming that at the k-th path point... , Then point This is the next tracking path point that the vehicle needs to follow.

[0037] In this embodiment, when selecting tracking path points, the vehicle's pose orientation vector relative to each parking path point is calculated. The inner product of the current vehicle pose and the pose orientation vector is calculated, and the first parking path point with an inner product greater than zero is obtained. This parking path point is selected from the parking path points that is consistent with the vehicle's current path direction. The difference between the parking path point and the vehicle's driving direction is identified by the vector inner product calculation, and reverse / unreachable path points are excluded. The driving direction switching point is automatically identified to ensure that the vehicle can be smoothly tracked at the selected starting point. While ensuring real-time performance, the directional robustness of the control system is significantly improved.

[0038] S3. Obtain the preceding and following path points adjacent to the tracking path point, and calculate the instantaneous tracking turning circular curve based on the preceding path point, the tracking path point, and the following path point, including: Based on the parking path point corresponding to the tracking path point, the previous parking path point adjacent to the tracking path point is obtained from the parking planning path as the preceding path point, and the next parking path point adjacent to the tracking path point is obtained from the parking planning path as the following path point. Substitute the preceding path point, the tracking path point, and the following path point into the general equation of a circle to obtain the instantaneous tracking turning circle curve, and determine the center coordinates and the turning circle radius.

[0039] Most existing path tracking algorithms eliminate lateral errors by controlling the front wheel steering angle, essentially acting as a P-controller based on lateral error. Since the vehicle system is an underactuated system (i.e., a system with fewer input variables than degrees of freedom), it is theoretically impossible to eliminate heading errors. Here, lateral error refers to the lateral error at the aiming point. Even if the lateral error is zero, it cannot guarantee that the lateral error at the front axle, rear axle, or center of gravity of the vehicle will be zero. Therefore, this embodiment uses path information before and after the path point (i.e., preceding path points and tracking path points) to eliminate lateral and heading errors during the path tracking process as much as possible.

[0040] See Figure 5 Step S2 calculates the next tracking path point that the vehicle needs to follow on the planned global path. Assuming... The next tracking path point to be tracked. and For the path point sequence and point Adjacent preorder and postorder path points.

[0041] Specifically: Suppose the general equation of a circle determined by three points is:

[0042] According to Cramer's Rule:

[0043]

[0044]

[0045]

[0046] The center coordinates are as follows The radius of the turning circle is denoted as ,but

[0047] During vehicle movement, since the parking planning path consists of discrete points, directly selecting the aiming point will result in calculation errors due to excessively large or small distances. To avoid excessive errors caused by the aiming point being too far from the vehicle, leading to increased lateral tracking errors and deviation from the parking planning path, as well as waste caused by small errors (the distance between points on the path and the vehicle; if the distance is too large, the calculation will be inaccurate; if the distance is too small, there will be no actual change in vehicle movement), a suitable aiming point is selected by instantaneously tracking the steering curve.

[0048] In this embodiment, the aiming distance is directly proportional to the vehicle's speed, and the aiming distance is determined in real time based on the vehicle's current speed.

[0049] Since the faster the vehicle speed, the greater the distance the vehicle travels per unit time, the selected aiming distance should be greater. Therefore, the aiming distance should be... The selection of the aiming distance should be proportional to the vehicle's speed v. Generally, the relationship between the aiming distance and the vehicle speed is selected as follows:

[0050] Where k is the aiming distance coefficient. The preset distance, the aiming distance coefficient, and the preset distance are all obtained through pre-calibration.

[0051] This embodiment calibrates the pre-aiming distance to be proportional to the vehicle's speed. At high speeds, the pre-aiming distance is increased to compensate for control delay, while at low speeds, the pre-aiming distance is shortened to improve accuracy. This achieves simultaneous improvement in delay compensation and stability, effectively increasing parking efficiency.

[0052] S4. Determining the aiming point from the tracking turning circular curve based on the aiming distance includes: Determine the position of the rear axle center point based on the vehicle body parameters; Extract a parking planning curve from the instantaneous tracking steering circular curve that lies between the preceding path point and the following path point; Draw a circle with the center point of the rear axle of the vehicle as the starting point and the aiming distance as the radius, and obtain the intersection point of the circle with the parking planning curve as the aiming point.

[0053] In this embodiment, when selecting the aiming point, a tracking steering circle curve is first drawn based on the parking path points (previous path points, tracking path points, and subsequent path points). Then, a circle is drawn with the rear axle center point of the vehicle as the starting point and the aiming distance as the radius. The intersection point of this circle with the parking planning curve is obtained as the aiming point. On the one hand, based on the parking path points, the tracking control of the parking planning path is realized as much as possible. On the other hand, dynamic selection based on the aiming distance fully matches the current vehicle posture (avoiding multiple adjustments), which can improve aiming accuracy, reduce the difficulty of tracking steering, and further realize path smoothing optimization.

[0054] S5. Calculate the target steering wheel angle based on the pre-aiming point for vehicle control, specifically: Based on the vehicle kinematics model, the current vehicle pose is determined, including the current heading angle and position coordinates; Obtain the pre-aiming coordinates of the pre-aiming point; The target steering wheel angle is calculated based on the current heading angle, position coordinates, aiming coordinates, and aiming distance, and vehicle control is performed based on the target steering wheel angle.

[0055] Specifically, based on the current position of the rear wheel center of the vehicle, a pre-aiming point is found on the reference path (the curved trajectory between parking path points on the parking planning path). It is assumed that the vehicle can drive to the target point with a certain turning radius. Then, the front wheel turning angle of the vehicle is calculated based on the geometric relationship between the distance from the current position of the vehicle to the pre-aiming point, the turning radius, and the angle between the pre-aiming point and the vehicle's front orientation, which controls the lateral tracking error of the vehicle.

[0056] See Figure 6 , Represents the distance from the rear axle center point M of the vehicle to The distance, i.e., the pre-aiming distance; The target point represents the current vehicle's aiming point, and is calculated by using the aiming distance as the radius. Rear axle center of vehicle The intersection of the circle with the center and the instantaneous tracking steering curve is obtained; The angle with the axle; This represents the lateral deviation between the aiming point and the vehicle; Represents wheelbase; This represents the target's turning radius during pure tracking. This represents the front wheel steering angle.

[0057] From the diagram Geometric relations can be obtained

[0058] After simplification, it becomes

[0059] From the simplified vehicle kinematics model and the geometric relationships in the figure above, we can obtain...

[0060] In this embodiment, the formula for calculating the target steering wheel angle is as follows:

[0061] In the formula, Indicates the target steering wheel angle. This indicates the aiming distance, and L represents the vehicle's wheelbase. Indicates the aiming distance The angle between the direction and the axle.

[0062] From the above formula, we can see that the control effect depends on the pre-aiming distance. The longer the aiming distance, the greater the front wheel steering angle. The smaller the change, the smoother the control effect; the shorter the aiming distance, the more precise the control effect, but it will also introduce some vibration. In summary, by sending the target steering wheel angle to the electronic power steering system, we can achieve steering wheel control of the vehicle.

[0063] This embodiment calculates the target steering wheel angle based on the current heading angle, position coordinates, preview coordinates, and preview distance, and controls the vehicle based on the target steering wheel angle, strictly following the bicycle kinematics model and effectively avoiding kinematic distortion.

[0064] Example 2 This invention also provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a parking control method based on path tracing as described in Embodiment 1 above.

[0065] In this embodiment of the invention, when performing parking control, a pre-planned parking path is used as the standard. Tracking path points are selected from the path based on the vehicle's pose, avoiding actuator oscillations caused by sudden changes in steering angle. Furthermore, an instantaneous tracking steering curve is calculated based on the tracking path points and adjacent preceding and following path points. Based on the pre-aiming distance, a pre-aiming point is determined from the tracking steering curve. By matching the curvature of the path points, pre-aiming points that meet kinematic constraints are selected to adapt to the constantly changing path curvature. On the other hand, tracking errors are effectively suppressed, and tracking calculation accuracy is improved. This improves parking efficiency in compact parking scenarios, enhances the parking experience, and better suits the driver's driving habits.

[0066] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A parking control method based on path tracking, characterized in that, include: A vehicle kinematics model is established, and a parking planning path is obtained; wherein, the parking planning path contains multiple discrete parking path points; The vehicle pose is determined based on the vehicle kinematics model, and tracking path points are selected from the parking path points of the parking planning path based on the vehicle pose. Obtain the preceding and following path points adjacent to the tracking path point, and calculate the instantaneous tracking turning circular curve based on the preceding path point, the tracking path point, and the following path point; The aiming point is determined from the tracking turning circular curve based on the aiming distance; The target steering wheel angle is calculated based on the pre-aiming point for vehicle control.

2. The parking control method based on path tracking as described in claim 1, characterized in that, The vehicle pose is determined based on the vehicle kinematic model, and tracking path points are selected from the parking path points of the parking planning path based on the vehicle pose, including: Obtain the path segment closest to the vehicle's current position from the parking planning path, and obtain all the parking path points on the path segment; Based on the vehicle kinematics model, the current vehicle pose is determined; Based on the vehicle kinematics model and the vehicle's current position, calculate the vehicle's pose orientation vector relative to each parking path point; Based on the pose orientation vector and the current vehicle pose, a tracking path point that is consistent with the current path direction of the vehicle is selected from the parking path points.

3. The parking control method based on path tracking as described in claim 2, characterized in that, Based on the pose orientation vector and the current vehicle pose, tracking path points that are consistent with the current path direction of the vehicle are selected from the parking path points, specifically as follows: Iterate through the current vehicle pose at each parking path point and calculate the inner product of the current vehicle pose and the pose orientation vector; iterate through the inner product of each parking path point, and when the first inner product greater than zero is detected, determine that the parking path point corresponding to this inner product is consistent with the current path direction of the vehicle, and determine the parking path point as the tracking path point.

4. The parking control method based on path tracking as described in claim 2, characterized in that, Based on the vehicle kinematics model and the vehicle's current position, the vehicle's pose orientation vector relative to each parking path point is calculated. Specifically, based on the vehicle kinematics model, the position coordinates of the vehicle's current position and the parking coordinates of each parking path point are determined. The pose orientation vector of the vehicle relative to each parking path point is calculated with the position coordinates as the starting point and the parking coordinates as the ending point.

5. The parking control method based on path tracking as described in claim 4, characterized in that, Obtaining the preceding and following path points adjacent to the tracking path point, and calculating the instantaneous tracking turning circular curve based on the preceding, following, and following path points, including: Based on the parking path point corresponding to the tracking path point, the previous parking path point adjacent to the tracking path point is obtained from the parking planning path as the preceding path point, and the next parking path point adjacent to the tracking path point is obtained from the parking planning path as the following path point. Substitute the preceding path point, the tracking path point, and the following path point into the general equation of a circle to obtain the instantaneous tracking turning circle curve, and determine the center coordinates and the turning circle radius.

6. The parking control method based on path tracking as described in claim 1, characterized in that, Determining the aiming point from the tracking steering circular curve based on the aiming distance includes: Determine the position of the rear axle center point based on the vehicle body parameters; Extract a parking planning curve from the instantaneous tracking steering circular curve that lies between the preceding path point and the following path point; Draw a circle with the center point of the rear axle of the vehicle as the starting point and the aiming distance as the radius, and obtain the intersection point of the circle with the parking planning curve as the aiming point.

7. The parking control method based on path tracking as described in claim 6, characterized in that: The aiming distance is directly proportional to the vehicle's speed, and is determined in real time based on the vehicle's current speed.

8. The parking control method based on path tracking as described in claim 1, characterized in that, The target steering wheel angle is calculated based on the pre-aiming point, specifically as follows: Based on the vehicle kinematics model, the current vehicle pose is determined, including the current heading angle and position coordinates; Obtain the pre-aiming coordinates of the pre-aiming point; The target steering wheel angle is calculated based on the current heading angle, position coordinates, aiming coordinates, and aiming distance, and vehicle control is performed based on the target steering wheel angle.

9. The parking control method based on path tracking as described in claim 8, characterized in that, The formula for calculating the target steering wheel angle is as follows: In the formula, Indicates the target steering wheel angle. This indicates the aiming distance, and L represents the vehicle's wheelbase. Indicates the aiming distance The angle between the direction and the axle.

10. A terminal device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the path-tracking-based parking control method as described in any one of claims 1 to 9.

Images