Path planning method, storage medium, product, equipment and vehicle

Abstract

The invention relates to a path planning method, a storage medium, a product, equipment and a vehicle, and the method comprises the steps: carrying out the dynamic filtering of first position data in a path planning result of a previous frame and second position data in an optimal path result of a current frame, and obtaining an inter-frame dynamic filtering result; wherein the inter-frame dynamic filtering result comprises position data; and performing path curvature smooth optimization on the inter-frame dynamic filtering result based on path pure tracking to obtain a path planning result. The smoothness and accuracy of path planning are improved, repeated shaking of the steering wheel is reduced, and driving safety is ensured.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and more particularly to a path planning method, storage medium, product, device, and vehicle. Background Technology

[0002] Intelligent driving technology is a crucial development direction in vehicle technology, and intelligent path planning is a core component of this technology. Currently, methods exist for secondary optimization of initial path planning data to improve its quality. However, the data after secondary optimization exhibits significant fluctuations. Using this optimized data for lateral vehicle control can easily lead to issues such as steering wheel vibration, impacting driving safety. Summary of the Invention

[0003] This application provides a path planning method, storage medium, product, device, and vehicle, which improves the smoothness and accuracy of path planning, reduces repeated steering wheel vibration, and ensures driving safety, thereby at least partially solving the above-mentioned technical problems.

[0004] To achieve the above objectives, according to a first aspect of this application, a path planning method is provided, comprising: performing dynamic filtering on first position data in the path planning result of the previous frame and second position data in the optimal path result of the current frame to obtain an inter-frame dynamic filtering result; wherein the inter-frame dynamic filtering result includes position data; and performing path curvature smoothing optimization on the inter-frame dynamic filtering result based on pure path tracking to obtain a path planning result.

[0005] Optionally, the location data includes lateral position and longitudinal position.

[0006] Optionally, the inter-frame dynamic filtering results may also include path curvature and path heading angle.

[0007] Optionally, the method further includes: performing secondary optimization on the initial path result of the current frame to obtain the optimal path result of the current frame.

[0008] Optionally, the step of dynamically filtering the first position data in the path planning result of the previous frame and the second position data in the optimal path result of the current frame to obtain the inter-frame dynamic filtering result includes: performing a weighted summation of the first position data in the path planning result of the previous frame and the second position data in the optimal path result of the current frame to obtain the inter-frame dynamic filtering result.

[0009] Optionally, the weight value of the first location data is positively correlated with the sampling distance; the weight value of the second location data is negatively correlated with the sampling distance; wherein, the sampling distance is the distance between the sampling point and the current position of the vehicle.

[0010] Optionally, the step of optimizing the path curvature smoothing of the inter-frame dynamic filtering results based on pure path tracking to obtain path planning results includes: adding the filtering results of predicted sampling points to the inter-frame dynamic filtering results to update the inter-frame dynamic filtering results; performing pure path tracking on the inter-frame dynamic filtering results to obtain path planning results; and removing the planning results of the predicted sampling points from the path planning results to update the path planning results.

[0011] Optionally, the method further includes: performing forward interpolation on the inter-frame dynamic filtering result to add the filtering result of forward prediction sampling points to the inter-frame dynamic filtering result; performing backward interpolation on the inter-frame dynamic filtering result to add the filtering result of backward prediction sampling points to the inter-frame dynamic filtering result; wherein the prediction sampling points include the forward prediction sampling points and the backward prediction sampling points.

[0012] Optionally, the filtering results of the forward prediction sampling points include location data, path curvature, and path heading angle.

[0013] Optionally, the filtering result of the backward prediction sampling point includes location data.

[0014] Optionally, the number of forward prediction sampling points is equal to the number of backward prediction sampling points.

[0015] Optionally, the step of performing pure path tracking on the inter-frame dynamic filtering results to obtain path planning results includes: determining a pre-aiming distance based on the inter-frame dynamic filtering results; and updating the inter-frame dynamic filtering results based on the pre-aiming distance to obtain path planning results.

[0016] Optionally, determining the aiming distance based on the inter-frame dynamic filtering result includes: determining the aiming curvature based on the inter-frame dynamic filtering result; determining the aiming speed based on the aiming curvature, the current vehicle speed, and the minimum protected vehicle speed; and determining the aiming distance based on the aiming speed, the preset aiming duration, the preset minimum aiming distance, and the preset maximum aiming distance.

[0017] Optionally, determining the aiming curvature based on the inter-frame dynamic filtering result includes: selecting the path curvature that satisfies the target condition from multiple path curvatures in the inter-frame dynamic filtering result as the aiming curvature.

[0018] Optionally, the sampling distance corresponding to multiple path curvatures is less than the sampling distance corresponding to the other path curvatures in the inter-frame dynamic filtering result; wherein, the sampling distance is the distance between the sampling point and the current position of the vehicle.

[0019] Optionally, the target condition includes maximizing the value of the path curvature.

[0020] Optionally, the method further includes: determining the target front wheel steering angle based on the path planning result.

[0021] Optionally, determining the target front wheel steering angle based on the path planning result includes: performing feedforward control processing on the path planning result to obtain a first front wheel steering angle; performing feedback control processing on the path planning result to obtain a second front wheel steering angle; and determining the target front wheel steering angle based on the first front wheel steering angle and the second front wheel steering angle.

[0022] According to a second aspect of this application, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described path planning method.

[0023] According to a third aspect of this application, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the path planning method described above.

[0024] According to a fourth aspect of this application, an electronic device is provided, comprising: a memory having a computer program stored thereon; and a processor for executing the computer program in the memory to implement the path planning method described above.

[0025] According to a fifth aspect of this application, a vehicle is provided, including the aforementioned electronic equipment.

[0026] The path planning method, storage medium, product, device, and vehicle provided in this application embodiment combine the path planning result of the previous frame with dynamic filtering of the optimal path result of the current frame to obtain an inter-frame dynamic filtering result; then, based on pure path tracking, the path curvature smoothing optimization of the inter-frame dynamic filtering result is performed to obtain the path planning result. Because the optimal path result of the current frame is dynamically filtered with reference to the path planning result of the previous frame, the driving path planning of the current frame can incorporate the driving path characteristics of historical frames. Furthermore, the path curvature smoothing optimization of the inter-frame dynamic filtering result based on pure path tracking allows the path planning result of the current frame to change smoothly, improving the smoothness and accuracy of path planning, thereby reducing repeated steering wheel vibrations and ensuring driving safety.

[0027] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of 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.

[0029] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0030] Figure 1 This is a flowchart of a path planning method provided in an embodiment of this application;

[0031] Figure 2 This is a schematic diagram of a dynamic filtering method provided in an embodiment of this application;

[0032] Figure 3 This is a schematic diagram of pure path tracing provided in an embodiment of this application;

[0033] Figure 4 This is a schematic diagram of the geometric relationship of a vehicle model provided in an embodiment of this application;

[0034] Figure 5 This is a schematic diagram of a lateral control process provided in an embodiment of this application;

[0035] Figure 6 This is a flowchart of another path planning method provided in an embodiment of this application;

[0036] Figure 7 This is a schematic diagram of a vehicle provided in an embodiment of this application. Detailed Implementation

[0037] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0038] According to a first aspect of this application, embodiments of this application provide a path planning method.

[0039] Please see Figure 1 , Figure 1 This is a flowchart illustrating a path planning method provided in an embodiment of this application. The path planning method may include the following steps:

[0040] Step S100: Perform dynamic filtering on the first position data in the path planning result of the previous frame and the second position data in the optimal path result of the current frame to obtain the inter-frame dynamic filtering result;

[0041] Step S200: Based on pure path tracking, perform path curvature smoothing optimization on the inter-frame dynamic filtering results to obtain the path planning results.

[0042] Since step S100 performs dynamic filtering on the first position data and the second position data, the resulting inter-frame dynamic filtering result includes position data. In some embodiments, the first position data in the path planning result of the previous frame, the second position data in the optimal path result of the current frame, and the position data in the inter-frame dynamic filtering result may respectively include horizontal and vertical positions, such as the horizontal and vertical positions of multiple sampling points.

[0043] The optimal path result refers to the path result obtained from path planning for the current frame. In some embodiments, before step S100, the path planning method further includes: performing secondary optimization on the initial path result of the current frame to obtain the optimal path result for the current frame. The initial path result refers to data obtained from preliminary planning of the vehicle's driving path. For example, the vehicle's decision-making module performs image recognition, path decision-making, and other processing through artificial intelligence algorithms to output the initial path result. Optionally, the initial path result includes, but is not limited to, at least one of the following: the initial lateral position of the sampling point, the initial longitudinal position of the sampling point, the initial path curvature of the sampling point, and the initial path heading angle of the sampling point. This application embodiment performs secondary optimization on the initial path result to improve the path planning quality. The data after secondary optimization of the initial path result is the optimal path result. Optionally, the optimal path result includes, but is not limited to, at least one of the following: the optimized lateral position of the sampling point, the optimized longitudinal position of the sampling point, the optimized path curvature of the sampling point, and the optimized path heading angle of the sampling point. The second position data in the optimal path result of the current frame may include the optimized lateral position and the optimized longitudinal position of the sampling point. For details on the secondary optimization process, please refer to the following examples, which will not be elaborated here.

[0044] It should be understood that in the embodiments of this application, if the current frame is the first frame, that is, if the current frame is the first frame sampled, the optimal path result of the current frame can be directly used as the path planning result of the current frame; if the current frame is not the first frame, that is, if there are other frames sampled before the current frame, then steps S100 and S200 are executed to obtain the path planning result of the current frame.

[0045] In summary, the path planning method provided in this application combines the path planning result of the previous frame with dynamic filtering of the optimal path result of the current frame to obtain an inter-frame dynamic filtering result; then, based on pure path tracking, path curvature smoothing optimization is performed on the inter-frame dynamic filtering result to obtain the path planning result. Because the optimal path result of the current frame is dynamically filtered with reference to the path planning result of the previous frame, the driving path planning of the current frame can incorporate the driving path characteristics of historical frames. Furthermore, the path curvature smoothing optimization based on pure path tracking of the inter-frame dynamic filtering result allows the path planning result of the current frame to change smoothly, improving the smoothness and accuracy of path planning, thereby reducing repeated steering wheel vibrations and ensuring driving safety.

[0046] In some embodiments, the path planning method described above may further include: establishing a quadratic optimization objective function and a quadratic optimization solver.

[0047] The secondary optimization solver is either the Ceres solver or the Corridor solver.

[0048] The inputs for secondary optimization include the initial path results, such as the initial longitudinal and lateral positions of the sampling points, the vehicle's speed, the path curvature, and the path heading angle.

[0049] The objective function for quadratic optimization can satisfy the constraint relationship shown in Equation 1 below:

[0050] Formula 1:

[0051] In the formula, y is an n-dimensional sampling point position matrix, and n is the number of sampling points; x T Let be the transpose of x, where x is a matrix composed of data from multiple sampling points, such as the lateral and longitudinal positions, lateral velocities, longitudinal velocities, lateral accelerations, longitudinal accelerations, path curvature, and path heading angles of the sampling points; y is the quadratic optimization objective function value; P is the loss function matrix; and q is the path heading angle. T This is the transpose of the position deviation matrix.

[0052] Among them, the sampling method of the sampling points can be equal-interval sampling. The sampling interval in equal-interval sampling is related to the current driving speed of the vehicle and can satisfy the relationship shown in Formula 2 below.

[0053] Formula 2: l = Δt·v ego

[0054] In the formula, l is the sampling interval, Δt is the sampling step size, and v ego This represents the vehicle's current speed.

[0055] Based on this, the optimal path result for the current frame includes the x values ​​of n sampling points when y reaches its minimum value in Formula 1, which are the optimized longitudinal position, lateral position, longitudinal velocity, lateral velocity, longitudinal acceleration, lateral acceleration, path curvature, and path heading angle of the sampling points. The optimal path result for the current frame is obtained under the condition of satisfying the constraints of the entire driving scenario. These constraints can satisfy the relationship shown in Formula 3. The entire driving scenario includes, but is not limited to, straight-line driving scenarios, small-curvature curve driving scenarios, large-curvature curve driving scenarios, S-curve driving scenarios, U-turn driving scenarios, lane-changing driving scenarios, and obstacle avoidance driving scenarios.

[0056] Formula 3: u low ≤A T x≤u up

[0057] In the formula, A T Let u be the transpose of the constraint matrix. low To constrain the lower boundary matrix, u up To constrain the upper boundary matrix.

[0058] In some embodiments, step S100 may include the following step S110:

[0059] Step S110: Perform a weighted summation of the first position data in the path planning result of the previous frame and the second position data in the optimal path result of the current frame to obtain the inter-frame dynamic filtering result.

[0060] Dynamic filtering involves a weighted summation of the first position data from the path planning result of the previous frame and the second position data from the optimal path result of the current frame. In some embodiments, both the first and second position data include data from multiple sampling points arranged according to sampling distance. This sampling distance is the distance between the sampling point and the vehicle's current position. In practical applications, sampling can be performed sequentially in ascending order of sampling distance, so that the order of sampling points in the path planning result, the optimal path result, and the inter-frame dynamic filtering result can also be consistent with the order of sampling times.

[0061] In some embodiments, both the first position data and the second position data include the longitudinal and / or lateral positions of the sampling points; thus, the position data in the inter-frame dynamic filtering result includes the filtered longitudinal and / or lateral positions of the sampling points. Embodiments of this application may perform dynamic filtering on the longitudinal and / or lateral positions of the sampling points, because the path formed after dynamic filtering of the path curvature and path heading angle differs significantly from the path formed after dynamic filtering of the longitudinal and lateral positions.

[0062] For example, such as Figure 2As shown, in this embodiment of the application, second position data is extracted from the optimal path result of the current frame. The second position data includes the vertical and horizontal positions after the sampling points are optimized. First position data is extracted from the path planning result of the previous frame. The first position data includes the vertical and horizontal positions after the sampling points are planned. Dynamic filtering is performed on the first position data and the second position data to obtain the inter-frame dynamic filtering result of the current frame. The position data in the inter-frame dynamic filtering result includes the vertical and horizontal positions after the sampling points are filtered.

[0063] It should be understood that, for ease of explanation, in the embodiments of this application, the path curvature and path heading angle in the inter-frame dynamic filtering result are referred to as the path curvature after sampling point filtering and the path heading angle after sampling point filtering, respectively. However, this does not mean that the path curvature and path heading angle need to be dynamically filtered. In the embodiments of this application, dynamic filtering can be applied only to the longitudinal and lateral positions, while the path curvature and path heading angle do not need to be dynamically filtered through weighted processing or other methods.

[0064] In some embodiments, the processing procedure of step S110 above can be as shown in Formula 4 below:

[0065] Formula 4:

[0066] Where, x i Let y be the filtered horizontal position of the i-th sampling point in the current frame. i x represents the filtered vertical position of the i-th sampling point in the current frame; i_cur y is the optimized horizontal position of the i-th sampling point in the current frame. i_cur x is the optimized vertical position of the i-th sampling point in the current frame; i_last Let y be the horizontal position of the i-th sampling point in the previous frame. i_last k1 represents the vertical position of the i-th sampling point in the previous frame after planning; k2 represents the weight value of the current frame and k1 represents the weight value of the previous frame.

[0067] In some embodiments, the weight value of the second position data in the current frame is negatively correlated with the sampling distance, while the weight value of the first position data in the previous frame is positively correlated with the sampling distance; where the sampling distance is the distance between the sampling point and the current position of the vehicle. That is, in Formula 4 above, k1 decreases as i increases, and k2 increases as i increases. When the sampling distance is smaller, the sampling point is closer to the vehicle, and the data from the current frame should be adopted more, thus reducing the weight value of the previous frame and increasing the weight value of the current frame; when the sampling distance is larger, the sampling point is farther from the vehicle, and the sampling point in the current frame at a distance may not be accurate, so the data from the previous frame should be adopted more to avoid repeated fluctuations in the vehicle position, thus increasing the weight value of the previous frame and decreasing the weight value of the current frame. The values ​​of k1 and k2 can be obtained by querying a preset mapping table; different mapping tables can be preset for different path curvatures or driving scenarios.

[0068] In summary, the path planning method provided in this application combines the path planning results of the previous frame with dynamic filtering of the optimal path results of the current frame. Moreover, this dynamic filtering is applied to the position data of the sampling points, rather than directly filtering the path curvature and path heading angle of the sampling points, thus ensuring the accuracy of the sampling points after filtering and avoiding distortion of the path planning results.

[0069] In some embodiments, step S200 above includes the following steps S121 to S123:

[0070] Step S210: Add the filtering results of the predicted sampling points to the inter-frame dynamic filtering results to update the inter-frame dynamic filtering results;

[0071] Step S220: Perform pure path tracking on the inter-frame dynamic filtering results to obtain the path planning results;

[0072] Step S230: Remove the planning results of the predicted sampling points from the path planning results to update the path planning results.

[0073] In this embodiment, the predicted sampling points are first expanded in the inter-frame dynamic filtering results to eliminate the initial spikes and sharp drops at the end of the pure path tracking process. Therefore, the predicted sampling points are a smoothing process quantity in the path planning results of pure path tracking. When outputting the path planning results, the planning results of the predicted sampling points need to be deleted from the path planning results to obtain the path planning results required for vehicle lateral control, etc.

[0074] Based on the analysis of the results of pure path tracking, the embodiments of this application show that there are large jumps in the position and path curvature of the first few sampling points in the path planning results of pure path tracking. Therefore, one or more predicted sampling points are introduced before the original sampling points to eliminate the spike problem in the pure path tracking results at the beginning of the original sampling points. At the same time, considering the pre-aiming characteristics of pure path tracking, if there is no pre-aiming data after the end of the original sampling points, the pure path tracking results at the end of the original sampling points will drop sharply. Therefore, one or more predicted sampling points are also introduced after the end of the original sampling points to solve the steep drop problem.

[0075] Based on this, in some embodiments, step S210 above may include the following steps S211 to S212:

[0076] Step S211: Perform forward interpolation on the inter-frame dynamic filtering results to add the filtering results of the forward prediction sampling points to the inter-frame dynamic filtering results;

[0077] Step S212: Perform backward interpolation on the inter-frame dynamic filtering results to add the filtering results of backward prediction sampling points to the inter-frame dynamic filtering results.

[0078] In some embodiments, based on considerations of data symmetry, the number of forward prediction sampling points can be equal to the number of backward prediction sampling points. Since the more prediction sampling points there are, the higher the data distortion rate at both ends of the original sampling points, the number of forward or backward prediction sampling points can be less than or equal to 5, for example, the number of forward or backward prediction sampling points is less than or equal to 3.

[0079] Since the path curvature and path heading angle of the first sampling point are used as the initial input for pure path tracking, interpolation calculations are performed on the path curvature and path heading angle during forward interpolation. In some embodiments, the filtered result of the forward prediction sampling point includes position data, path curvature, and path heading angle. For example, the filtered result of the forward prediction sampling point includes the filtered longitudinal position, lateral position, path curvature, and / or path heading angle of the sampling point. Forward interpolation can employ forward linear interpolation, thus step S211 can be as shown in Formula 5 below:

[0080] Formula 5:

[0081] Where, x i To predict the filtered lateral position of the |i|th sampling point, x1 represents the filtered lateral position of the first sampling point in the inter-frame dynamic filtering result, and x2 represents the filtered lateral position of the second sampling point in the inter-frame dynamic filtering result; y iTo predict the filtered vertical position of the |i|th sampling point, y1 represents the filtered vertical position of the first sampling point in the inter-frame dynamic filtering result, and y2 represents the filtered vertical position of the second sampling point in the inter-frame dynamic filtering result; k i To predict the path curvature after filtering at the |i|th sampling point, k1 is the path curvature after filtering at the first sampling point in the inter-frame dynamic filtering result, and k2 is the path curvature after filtering at the second sampling point in the inter-frame dynamic filtering result; ψ i To predict the path heading angle after filtering at the |i|th sampling point, ψ1 is the path heading angle after filtering at the first sampling point in the inter-frame dynamic filtering result, ψ2 is the path heading angle after filtering at the second sampling point in the inter-frame dynamic filtering result; m is the number of forward prediction sampling points.

[0082] Since it is not necessary to calculate the difference between path curvature and path heading angle during backward linear interpolation, in some embodiments, the filtered result of the backward prediction sampling point includes position data. For example, the filtered result of the backward prediction sampling point includes the filtered longitudinal position and / or lateral position of the sampling point. The backward interpolation can employ backward-limited interpolation, so step S212 can be as shown in Formula 6 below:

[0083] Formula 6:

[0084] Where, x i To predict the filtered lateral position of the i-th sampling point, x n x represents the horizontal position of the nth sampling point in the inter-frame dynamic filtering result after filtering. n-1 y represents the horizontal position of the (n-1)th sampling point in the inter-frame dynamic filtering result after filtering; i To predict the filtered vertical position of the i-th sampling point, y n y represents the vertical position of the nth sampling point in the inter-frame dynamic filtering result after filtering. n-1 is the vertical position of the (n-1)th sampling point in the inter-frame dynamic filtering result after filtering; m is the number of backward prediction sampling points.

[0085] For example, such as Figure 3 As shown, after dynamic filtering, the inter-frame dynamic filtering result is obtained. Then, the sampling points of the inter-frame dynamic filtering result are expanded by forward interpolation by m forward prediction sampling points and backward interpolation by m backward prediction sampling points. Forward interpolation needs to determine the lateral position, longitudinal position, path curvature and path heading angle of each prediction sampling point after filtering, and backward interpolation needs to determine the lateral position and longitudinal position of each sampling point after filtering.

[0086] In summary, the path planning method provided in this application expands the sampling points of the inter-frame dynamic filtering results after dynamic filtering. Specifically, by introducing one or more predicted sampling points before the original sampling points of the inter-frame dynamic filtering results through forward interpolation, the spike problem of the path planning results of subsequent pure path tracking can be eliminated; by introducing one or more predicted sampling points after the original sampling points of the inter-frame dynamic filtering results through backward interpolation, the steep drop problem of the path planning results of subsequent pure path tracking can be eliminated.

[0087] In some embodiments, step S220 may include the following steps S221 to S222:

[0088] Step S221: Determine the aiming distance based on the inter-frame dynamic filtering results;

[0089] Step S222: Update the inter-frame dynamic filtering results based on the pre-aiming distance to obtain the path planning results.

[0090] When performing pure path tracking, it is necessary to determine the aiming distance and then update the data of each sampling point in the inter-frame dynamic filtering result based on the aiming distance. This includes updating the path curvature, path heading angle, lateral position, and longitudinal position of the sampled points after filtering, in order to obtain the path planning result.

[0091] In some embodiments, step S221 above may include the following steps:

[0092] Step S2211: Determine the aiming curvature based on the inter-frame dynamic filtering results;

[0093] Step S2212: Determine the aiming speed based on the aiming curvature, the vehicle's current speed, and the vehicle's minimum protected speed;

[0094] Step S2213: Determine the aiming distance based on the aiming speed, preset aiming duration, preset minimum aiming distance, and preset maximum aiming distance.

[0095] The aiming distance is affected by the aiming vehicle speed, which in turn is affected by the aiming curvature. Therefore, in this embodiment, the aiming curvature is calculated first, the aiming vehicle speed is calculated based on the aiming curvature, and then the aiming distance is calculated based on the aiming vehicle speed.

[0096] In some embodiments, step S2211 may include: selecting a path curvature that satisfies a target condition from multiple path curvatures in the inter-frame dynamic filtering result as the pre-aiming curvature. The inter-frame dynamic filtering result includes multiple path curvatures. In some embodiments, the sampling distance corresponding to the multiple path curvatures is less than the sampling distance corresponding to the remaining path curvatures in the inter-frame dynamic filtering result; wherein the sampling distance is the distance between the sampling point and the current position of the vehicle. In some embodiments, the target condition is that the path curvature has the largest value. That is, the inter-frame dynamic filtering result may include multiple path curvatures, and in this embodiment, the path curvature with the largest value can be selected as the pre-aiming curvature from the first j path curvatures. The calculation method for j can be as shown in Formula 7 below:

[0097] Formula 7: j = min(t) k / Δt,m+n)

[0098] Among them, t k To preset the aiming time, t in Formula 7 k Round Δt to the nearest integer; Δt is the sampling step size; m refers to the number of forward prediction sampling points, and n refers to the number of sampling points in the inter-frame dynamic filtering result before sampling point expansion. Since the backward prediction sampling points do not have the path curvature after the prediction sampling points are filtered, the maximum value of j does not exceed m+n.

[0099] In some embodiments, step S2212 can be as shown in Formula 8 below:

[0100] Formula 8: v preview =max[min(v ego v table ), v min ]

[0101] Among them, v preview To anticipate vehicle speed; v ego The vehicle's current speed; v table The speed of the vehicle after the curvature speed limit; v min This is the minimum protected speed for the vehicle. The speed after curvature limiting is obtained by looking up the pre-aiming curvature in a table; the larger the pre-aiming curvature, the lower the speed after curvature limiting.

[0102] In some embodiments, step S2213 can be as shown in Formula 9 below:

[0103] Formula 9: x preview =min[max(t·v preview x min ), x max ]

[0104] Where, x preview This is the aiming distance; vpreview The target speed is t; the preset target time is x. min The preset minimum aiming distance; x max This is the preset maximum aiming distance.

[0105] For example, such as Figure 3 As shown, after expanding the sampling points, it is necessary to determine the dynamic preview distance for pure path tracking, including obtaining the preview curvature, obtaining the vehicle speed after the curvature speed limit, calculating the preview vehicle speed, and setting the minimum / maximum preview distance, so as to output the dynamic preview distance.

[0106] The longitudinal position, lateral position, path curvature, and path heading angle of the first sampling point in the path planning result are the same as those of the first sampling point in the inter-frame dynamic filtering result after filtering. Therefore, in step S222 above, the inter-frame dynamic filtering result is updated according to the pre-aiming distance, including the process of cyclically executing single-point path pure tracking starting from the second sampling point of the inter-frame dynamic filtering result.

[0107] First, such as Figure 3 As shown, it is necessary to determine the sampling point corresponding to the current location of the vehicle.

[0108] Taking the example of expanding the inter-frame dynamic filtering result by m prediction sampling points before and after, we traverse the n+2m sampling points in the inter-frame dynamic filtering result and calculate the distance between each sampling point in the inter-frame dynamic filtering result and the current position of the vehicle. The sampling point corresponding to the minimum distance is the sampling point corresponding to the current position of the vehicle.

[0109] Secondly, such as Figure 3 As shown, it is necessary to determine the sampling point corresponding to the aiming point position.

[0110] Starting from the sampling point corresponding to the vehicle's current position, traverse the sampling points in the inter-frame dynamic filtering results. When the distance between the vehicle and the pre-aiming point is greater than or equal to the pre-aiming distance for the first time, the sampling point traversed at this time is the sampling point corresponding to the pre-aiming point position. The distance between the vehicle and the pre-aiming point can be shown in Formula 10 below:

[0111] Formula 10:

[0112] Among them, l preview i is the distance between the vehicle and the target point; i0 is the sampling point corresponding to the current position of the vehicle; x i x represents the horizontal position of the i-th sampling point in the inter-frame dynamic filtering result. i+1 y represents the horizontal position of the (i+1)th sampling point in the inter-frame dynamic filtering result; i Let y be the vertical position of the i-th sampling point in the inter-frame dynamic filtering result. i+1This represents the vertical position of the (i+1)th sampling point in the inter-frame dynamic filtering result.

[0113] Then, as Figure 3 As shown, the path curvature and path heading angle need to be determined.

[0114] See Figure 4 Assuming the vehicle model is a two-wheeled vehicle centered on the rear axle, according to the geometric relationship of the vehicle model, the front wheel steering angle of the vehicle satisfies the relationship shown in Formula 11 below:

[0115] Formula 11: δ=atan[2·L·sin(α) / x preview ]

[0116] Where δ is the front wheel steering angle of the vehicle, L is the wheelbase of the vehicle, α is the angle between the line connecting the rear axle center of the vehicle and the sampling point corresponding to the preview point position and the vehicle body direction, and x preview This is the pre-aiming distance.

[0117] In Formula 11, α satisfies the relationship shown in Formula 12:

[0118] Formula 12: α = asin(e / l0)

[0119] Where e is the lateral position deviation between the rear axle center of the vehicle and the sampling point corresponding to the preview point, and l0 is the distance between the rear axle center of the vehicle and the sampling point corresponding to the preview point.

[0120] Combining formulas 11 and 12 above, we can obtain formula 13:

[0121] Formula 13:

[0122] Based on the vehicle model, by controlling δ to make e converge to a preset range, when e is adjusted to the preset range, the angle between the line connecting the rear axle center of the vehicle and the sampling point corresponding to the pre-aiming point position and the vehicle body direction is α0, and the front wheel rotation angle of the vehicle is δ0.

[0123] Based on formulas 14 and 15, the path curvature of the current sampling point in the path planning result can be calculated as shown in formula 16.

[0124] Formula 14: R=l0 / [2·sin(α0)]

[0125] Where R is the path radius of the current sampling point in the path planning result.

[0126] Formula 15: k = 1 / R

[0127] Where k is the path curvature of the current sampling point in the path planning result.

[0128] Formula 16: k=2·sin(α0) / l0

[0129] According to Formula 17, the path heading angle can be calculated as shown in Formula 18.

[0130] Formula 17:

[0131] in, v represents the rate of change of the path heading angle at the current sampling point in the path planning results. ego L represents the vehicle's current speed, and L represents the vehicle's wheelbase.

[0132] Formula 18:

[0133] In the formula, ψ is the path heading angle of the current sampling point in the path planning result, ψ0 is the path heading angle of the previous sampling point in the path planning result, and Δt is the sampling step size.

[0134] After that, as Figure 3 As shown, the remaining data in the path planning results are updated.

[0135] The vehicle's current position is updated according to Formula 19, and the updated current position is used as the current sampling point position in the path planning result. The current sampling point position in the path planning result includes the lateral position and the longitudinal position. The longitudinal velocity and longitudinal acceleration of the current sampling point in the path planning result are calculated based on the longitudinal position of the current sampling point in the path planning result. The lateral velocity and lateral acceleration of the current sampling point in the path planning result are calculated based on the lateral position of the current sampling point in the path planning result.

[0136] Formula 19:

[0137] In the formula, x update y represents the lateral position of the sampling point in the path planning result. update x represents the longitudinal position of the sampling point in the path planning result. last y represents the lateral position of the previous sampling point in the path planning result. last v represents the longitudinal position of the previous sampling point in the path planning result. ego ψ represents the vehicle's current speed, ψ represents the path heading angle of the current sampling point in the path planning results, and Δt represents the sampling step size.

[0138] Using the above method, the longitudinal position, lateral position, longitudinal velocity, lateral velocity, longitudinal acceleration, lateral acceleration, path curvature, and path heading angle of n+2m sampling points are collectively used as the path planning result, which is the path planning result obtained in step S220. In step S230, the planning results of the predicted sampling points are deleted from the path planning result, that is, the planning results of m forward predicted sampling points and m backward predicted sampling points, in order to update the path planning result, as follows. Figure 3 As shown, subsequent actions such as lateral vehicle control can be performed based on the updated path planning results.

[0139] In summary, the path planning method provided in this application reduces spikes and fluctuations in path curvature and heading angle through pure path tracking. Furthermore, compared to applying secondary optimization to the filtering results, this application combines dynamic filtering and pure path tracking after secondary optimization, requiring only one secondary optimization, resulting in lower computational cost and shorter computation time. In addition, the pure path tracking in this application is based on dynamic pre-aiming distance, which can determine a smoother path curvature for curves with large curvature and curvature change rates, reducing the error with the actual road curvature and minimizing internal and external tangent phenomena.

[0140] In some embodiments, the above path planning method may further include the following step S300:

[0141] Step S300: Determine the target front wheel steering angle based on the path planning results.

[0142] In the lateral control of a vehicle, the target steering angle of the front wheels can be calculated based on the path planning results after dynamic filtering and pure path tracking.

[0143] In some embodiments, step S300 above may include the following steps:

[0144] Step S310: Perform feedforward control processing on the path planning results to obtain the first front wheel steering angle;

[0145] Step S320: Perform feedback control processing on the path planning results to obtain the second front wheel steering angle;

[0146] Step S330: Determine the target front wheel angle based on the first front wheel angle and the second front wheel angle.

[0147] Lateral control includes lateral feedforward control and lateral feedback control. Lateral feedforward control performs feedforward control processing to obtain the vehicle's first front wheel steering angle, while lateral feedback control performs feedback control processing to obtain the vehicle's second front wheel steering angle. Based on the first and second front wheel steering angles, the target front wheel steering angle can be obtained by summing them. At each moment, by controlling the vehicle's front wheel steering angle to reach the target angle, the vehicle can be controlled to travel along the optimal driving path planned by the path planning results.

[0148] The feedforward control processing can process the path curvature of the sampling points in the path planning results and the vehicle's current speed; the feedback control processing can process the path heading angle and lateral position of the sampling points in the path planning results and the vehicle's current lateral position.

[0149] For example, such as Figure 5 As shown, the path curvature of the sampling points is extracted from the path planning results and combined with the vehicle's current speed as the input for lateral feedforward control to obtain the first front wheel steering angle. The lateral position of the sampling points is extracted from the path planning results and combined with the vehicle's current lateral position to determine the lateral position error of the sampling points. The path heading angle of the sampling points extracted from the path planning results is then combined with the path heading angle of the sampling points as the input for lateral feedback control to obtain the second front wheel steering angle. Finally, the first and second front wheel steering angles are summed to obtain the target front wheel steering angle.

[0150] In summary, the path planning method provided in this application determines the target front wheel steering angle of the vehicle based on the path planning results after dynamic filtering and pure path tracking. This allows the vehicle to be controlled to reach the target front wheel steering angle at each moment, thereby controlling the vehicle to travel along the optimal driving path planned by the path planning results. Because dynamic filtering and pure path tracking reduce the sharpness and fluctuations in the path curvature and heading angle obtained from secondary optimization, lateral control can achieve smoother angle control, reducing repeated steering wheel vibration and ensuring driving safety.

[0151] The path planning method provided in this application embodiment will be described below with an example.

[0152] Please see Figure 6 , Figure 6 This is a flowchart of a path planning method provided in an embodiment of this application. The path planning method may include the following steps S610 to S650:

[0153] Step S610: Perform secondary optimization on the initial path result of the current frame to obtain the optimal path result of the current frame;

[0154] Step S620: Determine whether the current frame is the first frame; if the current frame is the first frame, proceed from step S650 below; if the current frame is not the first frame, proceed from step S630 below.

[0155] Step S630: Perform dynamic filtering on the first position data in the path planning result of the previous frame and the second position data in the optimal path result of the current frame to obtain the inter-frame dynamic filtering result;

[0156] Step S640: Smooth the inter-frame dynamic filtering results based on pure path tracking to obtain the path planning results; wherein, the path planning results can be used as a reference for dynamic filtering of the optimal path results in the next frame;

[0157] Step S650: Perform lateral control processing of the vehicle based on the path planning result; wherein, if the current frame is the first frame, the path planning result of the current frame is the optimal path result of the current frame.

[0158] After obtaining the path planning result of the current frame through step S640, the path planning result of the current frame can be saved. Thus, when determining the path planning result of the next frame, the path planning result of the current frame can be used as a reference for dynamic filtering of the optimal path result of the next frame.

[0159] related Figure 6 For a description of each step in the embodiments and its beneficial effects, please refer to the above. Figures 1 to 5 The steps and their beneficial effects described in the embodiments are not elaborated here.

[0160] According to a second aspect of this application, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the path planning method described above. This non-transitory computer-readable storage medium possesses all the beneficial effects of the path planning method described above, which will not be elaborated further here.

[0161] According to a third aspect of this application, embodiments of this application also provide an electronic device, including: a memory and a processor, wherein a computer program is stored in the memory; the processor is configured to execute the computer program in the memory to implement the steps of the path planning method described above. This electronic device possesses all the beneficial effects of the path planning method described above, which will not be elaborated further here.

[0162] According to a fourth aspect of this application, embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the above-described path planning method.

[0163] Computer-readable storage media can be, for example, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof, without particular limitation herein. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0164] In some embodiments of this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that may be used or combined with an instruction execution system, apparatus, or device.

[0165] The aforementioned computer-readable storage medium may be included in the aforementioned electronic device, or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable storage medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0166] Dynamic filtering is performed on the first position data in the path planning result of the previous frame and the second position data in the optimal path result of the current frame to obtain the inter-frame dynamic filtering result; wherein, the inter-frame dynamic filtering result includes position data;

[0167] Based on pure path tracking, the inter-frame dynamic filtering results are optimized for path curvature smoothing to obtain path planning results.

[0168] Computer program code for performing operations of some embodiments of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0169] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function.

[0170] It should also be noted that in some alternative implementations, the functions marked in the box may occur in a different order than those marked in the attached figures.

[0171] For example, two consecutively represented blocks can actually be executed in substantially parallel order, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, as well as combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0172] The units described in some embodiments of this application can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a path filtering module and a path optimization module. The names of these units do not necessarily limit the specific unit; for example, the path optimization module may also be described as "a unit for performing path curvature smoothing optimization on the inter-frame dynamic filtering results based on pure path tracking to obtain path planning results."

[0173] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0174] According to the fifth aspect of this application, such as Figure 7 As shown in the illustration, this application also provides a vehicle 10, which includes the aforementioned electronic equipment. This vehicle possesses all the beneficial effects of the aforementioned electronic equipment, etc., which will not be elaborated upon further herein.

[0175] The vehicle may be a gasoline-powered vehicle, a plug-in hybrid electric vehicle, or a new energy vehicle, etc., and this application does not make any specific restrictions.

[0176] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0177] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0178] The embodiments, implementation methods, and related technical features of this application can be combined and substituted for each other without conflict.

[0179] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Although the descriptions of each embodiment in this application have different focuses, and parts not described in detail in a certain embodiment can be referred to the relevant descriptions of other embodiments, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of this application without departing from the content of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A path planning method characterized by, Comprising: performing dynamic filtering on first position data in a previous frame path planning result and second position data in a current frame optimal path result to obtain an inter-frame dynamic filtering result; wherein the inter-frame dynamic filtering result comprises position data; performing path curvature smoothing optimization on the inter-frame dynamic filtering result based on path pure tracking to obtain a path planning result.

2. The method of claim 1, wherein, The position data comprises lateral position and longitudinal position.

3. The method of claim 1, wherein, The inter-frame dynamic filtering result further comprises path curvature and path heading angle.

4. The method of claim 1, wherein, The method further comprises: performing secondary optimization on a current frame initial path result to obtain a current frame optimal path result.

5. The method of claim 1, wherein, The performing dynamic filtering on first position data in a previous frame path planning result and second position data in a current frame optimal path result to obtain an inter-frame dynamic filtering result comprises: performing weighted summation on first position data in a previous frame path planning result and second position data in a current frame optimal path result to obtain an inter-frame dynamic filtering result.

6. The method of claim 5, wherein: a weight value of the first position data is positively correlated with a sampling distance; a weight value of the second position data is negatively correlated with the sampling distance; wherein the sampling distance is a distance between a sampling point and a current position of the vehicle.

7. The method of claim 1, wherein, The performing path curvature smoothing optimization on the inter-frame dynamic filtering result based on path pure tracking to obtain a path planning result comprises: adding filtering results of predicted sampling points in the inter-frame dynamic filtering result to update the inter-frame dynamic filtering result; performing path pure tracking on the inter-frame dynamic filtering result to obtain a path planning result; eliminating planning results of the predicted sampling points from the path planning result to update the path planning result.

8. The method of claim 7, wherein, The method further comprises: performing forward interpolation on the inter-frame dynamic filtering result to add filtering results of forward predicted sampling points in the inter-frame dynamic filtering result; performing backward interpolation on the inter-frame dynamic filtering result to add filtering results of backward predicted sampling points in the inter-frame dynamic filtering result; wherein the predicted sampling points comprise the forward predicted sampling points and the backward predicted sampling points.

9. The method of claim 8, wherein, The filtering results of the forward predicted sampling points comprise position data, path curvature and path heading angle.

10. The method of claim 8, wherein, The filtering results of the backward predicted sampling points comprise position data.

11. The method of claim 8, wherein, The number of the forward predicted sampling points is equal to the number of the backward predicted sampling points.

12. The method of claim 7, wherein, The performing path pure tracking on the inter-frame dynamic filtering result to obtain a path planning result comprises: determining a preview distance according to the inter-frame dynamic filtering result; updating the inter-frame dynamic filtering result according to the preview distance to obtain a path planning result.

13. The method of claim 12, wherein, The determining a preview distance according to the inter-frame dynamic filtering result comprises: determining a preview curvature according to the inter-frame dynamic filtering result; determining a preview speed according to the preview curvature, a current vehicle speed and a minimum protection vehicle speed of the vehicle; determining a preview distance according to the preview speed, a preset preview time length, a preset minimum preview distance and a preset maximum preview distance.

14. The method of claim 13, wherein, The determining a preview curvature according to the inter-frame dynamic filtering result comprises: From the path curvatures in the inter-frame dynamic filtering result, a path curvature satisfying a target condition is selected as a preview curvature.

15. The method of claim 14, wherein, The path curvatures corresponding to a plurality of sampling distances are less than the path curvatures corresponding to the remaining sampling distances in the inter-frame dynamic filtering result; wherein the sampling distance is the distance between the sampling point and the current position of the vehicle.

16. The method of claim 14, wherein, The target condition includes that the path curvature has the maximum value.

17. The method of claim 1, wherein, The method further comprises: According to the path planning result, a target front wheel steering angle is determined.

18. The method of claim 17, wherein, According to the path planning result, a target front wheel steering angle is determined. The path planning result is subjected to feedforward control processing to obtain a first front wheel steering angle; The path planning result is subjected to feedback control processing to obtain a second front wheel steering angle; According to the first front wheel steering angle and the second front wheel steering angle, a target front wheel steering angle is determined.

19. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by a processor to implement the path planning method of any one of claims 1-18.

20. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the path planning method of any one of claims 1-18.

21. An electronic device, comprising: Comprise: A memory having a computer program stored thereon; A processor configured to execute the computer program in the memory to implement the path planning method of any one of claims 1-18.

22. A vehicle characterized by The electronic device of claim 21 is included.

Images