Vehicle trajectory planning method and device, electronic equipment and storage medium

Abstract

Embodiments of the present disclosure disclose a vehicle trajectory planning method and device, electronic equipment and storage medium, the method comprising: obtaining an initial reference trajectory of a target vehicle in a region to be planned, taking the maximum average degree of the projection distribution of the target vehicle on the plane where the initial reference trajectory is located on both sides of the trajectory as the objective function, and taking the kinematic model and safety constraints as the limiting conditions, sequentially performing quadratic programming to obtain a target control quantity sequence including the actual steering curvature corresponding to each sampling point, and then determining the target reference trajectory and each tracking point in the driving process of the target reference trajectory through the target control quantity sequence, avoiding the problem that when the target vehicle drives along the target reference trajectory, the distance of one side of the vehicle relative to the trajectory is too large, causing the side to be too close to the road boundary or roadside obstacles in the region, thereby solving the problem that in the prior art, it is difficult to plan a trajectory for a long-wheelbase vehicle to safely pass through a narrow region.

Description

Technical Field

[0001] This disclosure relates to the field of autonomous driving technology, and in particular to a vehicle trajectory planning method, apparatus, electronic device, and storage medium. Background Technology

[0002] As autonomous driving technology matures, the application of driverless vehicles is becoming increasingly widespread. Besides conventional or small vehicles like Robotaxi and Robo-delivery, large and medium-sized vehicles such as driverless buses, intelligent container transport vehicles, container trucks, and driverless heavy trucks are also attracting significant attention. These vehicles are characterized by longer wheelbases and wider bodies, requiring more turning space when navigating curves. The application of driverless vehicles in scenarios such as residential areas and construction diversion sections is also gradually expanding. These scenarios are characterized by narrow spaces and significant road curvature in certain areas.

[0003] Existing trajectory planning methods for autonomous vehicles are typically designed for vehicles with standard wheelbases and use fixed tracking points for trajectory planning. Common methods include using the rear axle center or center of gravity as the tracking point. However, this approach is difficult to apply to long-wheelbase vehicles, especially in confined spaces, where it fails to plan a feasible trajectory for safe passage through such areas. Summary of the Invention

[0004] To address, or at least partially address, the aforementioned technical problems, this disclosure provides a vehicle trajectory planning method, apparatus, electronic device, and storage medium, resolving the issue in the prior art where it is difficult to plan a trajectory for long-wheelbase vehicles to safely pass through narrow areas.

[0005] In a first aspect, embodiments of this disclosure provide a vehicle trajectory planning method, the method comprising:

[0006] Obtain the initial reference trajectory of the target vehicle within the area to be planned, wherein the initial reference trajectory includes each sampling point;

[0007] Using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point;

[0008] Based on the target control quantity sequence, a target reference trajectory is determined, and each tracking point is determined to minimize the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory. The tracking points are the position points on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same.

[0009] The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances on both sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point.

[0010] Secondly, embodiments of this disclosure also provide a vehicle trajectory planning device, the device comprising:

[0011] An initial trajectory determination module is used to obtain an initial reference trajectory of the target vehicle in the area to be planned, wherein the initial reference trajectory includes each sampling point;

[0012] The sequential quadratic programming module is used to perform sequential quadratic programming with the objective function of maximizing the uniformity of the projection distribution of the initial reference trajectory and the kinematic model and safety constraints of the target vehicle as the limiting conditions, to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point;

[0013] The target trajectory determination module is used to determine a target reference trajectory based on the target control quantity sequence, and to determine each tracking point that minimizes the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory, wherein the tracking point is a position point on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same;

[0014] The projection distribution uniformity is the average degree to which the projection of the target vehicle onto the plane of the initial reference trajectory is distributed on both sides of the trajectory. It reflects the difference in distance between the two sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of each sampling point is determined based on the kinematic model, the state vectors of adjacent sampling points, and the actual steering curvature of the adjacent sampling points.

[0015] Thirdly, embodiments of this disclosure also provide an electronic device, the electronic device comprising: one or more processors; a storage device for storing one or more programs; and when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the vehicle trajectory planning method as described above.

[0016] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the vehicle trajectory planning method as described above.

[0017] This disclosure provides a vehicle trajectory planning method. It obtains an initial reference trajectory of the target vehicle within a planned area. The objective function is to maximize the average distribution of the target vehicle's projection onto the plane of the initial reference trajectory on both sides of the trajectory. Using a kinematic model and safety constraints as limitations, sequential secondary planning is performed to obtain a sequence of target control quantities, including the actual steering curvature corresponding to each sampling point. This sequence of target control quantities is then used to determine the target reference trajectory and various tracking points that are not entirely identical during the journey along the target reference trajectory. The goal is to obtain the target reference trajectory and the tracking points that minimize the difference in distance between the two sides of the vehicle relative to the trajectory during the journey. This avoids the problem of excessive distance on one side of the vehicle relative to the trajectory when the target vehicle is traveling along the target reference trajectory, which could lead to a collision risk if that side is too close to the road boundary or roadside obstacle within the area. This ensures the vehicle can safely pass through narrow areas and solves the problem in the prior art where it is difficult to plan a safe trajectory for long-wheelbase vehicles to pass through narrow areas. Attached Figure Description

[0018] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0019] Figure 1 This is a schematic diagram of trajectory planning in the prior art according to an embodiment of this disclosure;

[0020] Figure 2 This is a schematic diagram illustrating a vehicle trajectory planning method according to an embodiment of this disclosure;

[0021] Figure 3 This is a flowchart of a vehicle trajectory planning method according to an embodiment of the present disclosure;

[0022] Figure 4 This is a schematic diagram of a vehicle lateral deviation in an embodiment of this disclosure;

[0023] Figure 5 This is an approximate schematic diagram of tracking deviation in the Frenet coordinate system according to an embodiment of this disclosure;

[0024] Figure 6 This is a schematic diagram of relevant parameters for a safety margin reference value in an embodiment of this disclosure;

[0025] Figure 7 This is a schematic diagram of the structure of a vehicle trajectory planning device according to an embodiment of the present disclosure;

[0026] Figure 8 This is a schematic diagram of the structure of an electronic device according to an embodiment of this disclosure. Detailed Implementation

[0027] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0028] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0029] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0030] Before providing a detailed description of the vehicle trajectory planning method provided in the embodiments of this disclosure, the technical problem solved by the vehicle trajectory planning method will be explained first. In the prior art, autonomous vehicle trajectory planning typically involves determining the trajectory and then using fixed tracking points to control the vehicle to travel along the trajectory, such as the rear axle center or the vehicle's center of gravity. The tracking point can be understood as a position on the vehicle that needs to travel along the trajectory.

[0031] However, this existing trajectory planning method, when using fixed tracking points to move along a trajectory, can cause the vehicle to deviate to one side of the road at some point, increasing the possibility of the vehicle going beyond the road boundary or colliding with roadside obstacles. For example, Figure 1 This is a schematic diagram of trajectory planning in the prior art according to an embodiment of this disclosure. Figure 1 It can be seen that when moving along a fixed trajectory using the rear axle center of the vehicle, there is a problem that the side of the vehicle body may be too close to the road boundary, leading to a collision.

[0032] Therefore, to solve the aforementioned technical problems and address the difficulty of long-wheelbase vehicles safely passing through narrow areas, this disclosure provides a vehicle trajectory planning method applicable to long-wheelbase vehicles, planning their driving trajectories in narrow areas. This method obtains a target reference trajectory and various tracking points along that trajectory, i.e., dynamically adjusted tracking points within the trajectory. When the vehicle travels along the trajectory using these tracking points, the difference in distance between the two sides of the vehicle relative to the trajectory is always minimized, preventing one side of the vehicle from getting too close to road boundaries or roadside obstacles, reducing the likelihood of the vehicle exceeding road boundaries or colliding with obstacles, thereby improving the trajectory's passability and safety. For example, Figure 2 This is a schematic diagram illustrating a vehicle trajectory planning method according to an embodiment of this disclosure. Figure 2 It can be seen that by determining the tracking point that minimizes the difference between the distances on both sides of the vehicle relative to the trajectory, the vehicle always maintains the maximum distance from the road boundary or roadside obstacles when using each tracking point to control the vehicle to travel along the trajectory, thereby improving the safety of the vehicle's passage.

[0033] Figure 3 This is a flowchart illustrating a vehicle trajectory planning method according to an embodiment of this disclosure. The method can be executed by a vehicle trajectory planning device, which can be implemented in software and / or hardware, and can be configured in an electronic device. Figure 3 As shown, the method may specifically include the following steps:

[0034] S110. Obtain the initial reference trajectory of the target vehicle in the area to be planned, wherein the initial reference trajectory includes each sampling point.

[0035] The target vehicle can be a vehicle currently on the planned trajectory. In this embodiment, the target vehicle can be a large or medium-sized vehicle with a long wheelbase and wide body, such as an unmanned bus, a container intelligent transfer vehicle, or an unmanned heavy truck. The area to be planned can be the area that the target vehicle needs to pass through, such as a narrow area. Specifically, a narrow area can be an area with a small passage width or a relatively large local road area.

[0036] In this embodiment, for a target vehicle whose trajectory needs to be planned, an initial reference trajectory for the target vehicle within the area to be planned can be determined first. Specifically, lane lines within the area to be planned can be determined, and then the lane lines can be used as the initial reference trajectory. It should be noted that the purpose of using lane lines as the initial reference trajectory is that, since lane lines are located at the center of the lane, driving along lane lines allows the vehicle to stay as close to the center of the lane as possible, thus staying as far away as possible from road boundaries or obstacles within the area to be planned.

[0037] Furthermore, after obtaining the initial reference trajectory, multiple sampling points can be determined within the initial reference trajectory. Each sampling point is located on the initial reference trajectory.

[0038] Specifically, starting from the target vehicle's current projected position on the initial reference trajectory and moving forward, multiple sampling points are obtained by sampling the initial reference trajectory according to a preset sampling distance. The current projected position can be a point on the initial reference trajectory mapped from the target vehicle's current position. Specifically, a line perpendicular to the initial reference trajectory can be drawn along the current position; the intersection of this line and the initial reference trajectory is the current projected position. The preset sampling distance can be pre-set or determined based on a preset number of samples and the length of the initial reference trajectory. The distance between two adjacent sampling points is equal to the preset sampling distance.

[0039] For example, if the preset sampling distance is Δs, and the preset number of samples forward from the current projection position is N, then the length of the initial reference trajectory is N*Δs, and N+1 sampling points can be obtained.

[0040] S120. Taking the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point.

[0041] Among them, the projection distribution uniformity is the average degree of the projection distribution of the target vehicle on the plane where the initial reference trajectory is located on both sides of the trajectory. It is used to reflect the difference between the distances on both sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point.

[0042] It should be noted that projection distribution uniformity can also be understood as the degree of similarity between the area of ​​the target vehicle's projection on one side of the initial reference trajectory and the area of ​​its projection on the other side. Specifically, the greater the projection distribution uniformity, the closer the area of ​​the target vehicle's projection on one side of the initial reference trajectory is to the area of ​​its projection on the other side; the smaller the difference between the distances of one side of the target vehicle relative to the initial reference trajectory and the distances of the other side of the target vehicle relative to the initial reference trajectory; the closer the two ends of the target vehicle that are furthest from the initial reference trajectory are to the initial reference trajectory; and the greater the distance between the two sides of the target vehicle and the road boundaries or obstacles of the area to be planned.

[0043] In this embodiment, the average degree of the projection distribution of the target vehicle on both sides of the initial reference trajectory can be determined based on the state vector of each sampling point. Then, the uniformity of the projection distribution of the initial reference trajectory can be determined based on the average degree of each sampling point.

[0044] The state vector at a sampling point can be a vector describing the actual state of the target vehicle at that sampling point, and can be determined based on the kinematic model, the actual steering curvature of the previous sampling point, and the state vector of the previous sampling point. The actual steering curvature at a sampling point can be the actual vehicle steering curvature of the target vehicle at that sampling point, and can be used as a control variable to determine the actual steering curvature of the next sampling point.

[0045] For example, the state vector can be the first lateral deviation of the rear axle center point of the target vehicle relative to the initial reference trajectory, and the kinematic model can be a mathematical model used to determine the state vector of each sampling point.

[0046] In one specific implementation, before performing sequential secondary planning, the method further includes: determining the arc transformation relationship between the first lateral deviation of the rear axle center point of the target vehicle relative to the initial reference trajectory, the heading deviation of the rear axle center point, and the second lateral deviation of the preset position point of the target vehicle relative to the initial reference trajectory; determining the first lateral deviation, the heading deviation, and the second lateral deviation as state vectors, and determining the state vector transformation relationship between the state vectors of adjacent sampling points; and constructing the kinematic model of the target vehicle based on the arc transformation relationship and the state vector transformation relationship.

[0047] The rear axle center point is the center of the rear axle on the target vehicle. The first lateral deviation of the rear axle center point relative to the initial reference trajectory can be the distance between the rear axle center point and its projection point on the initial reference trajectory. The heading deviation of the rear axle center point can specifically be the angle between the tangent line of the projection point on the initial reference trajectory and the vehicle's central axis. The projection point can be the location point on the initial reference trajectory projected from the rear axle center point; for example, the intersection of a straight line perpendicular to the initial reference trajectory drawn along the rear axle center point and the initial reference trajectory can be used as the projection point.

[0048] The preset position point can be a point located on the central axis of the target vehicle, and whose distance from the center point of the rear axle is not less than a set value; wherein, the set value can be determined according to the vehicle wheelbase. This application does not limit the setting position of the preset position point. In this embodiment, the preset position point and the center point of the rear axle are related to the state vector describing the actual state of the target vehicle, and the state vector determines the uniformity of the projection distribution. Therefore, the preset position point and the center point of the rear axle can be understood as position points for measuring the uniformity of the projection distribution, which affect the accuracy of the uniformity of the projection distribution.

[0049] For example, a point symmetrical to the center point of the rear axle can be selected on the vehicle's central axis as a preset position point, i.e., the center point of the front axle can be used as the preset position point to ensure the accuracy of the calculated uniformity of the projection distribution. The second lateral deviation of the preset position point relative to the initial reference trajectory can be the second lateral deviation of the center point of the front axle relative to the initial reference trajectory, that is, the distance between the center point of the front axle and the projection point of the center point of the front axle on the initial reference trajectory.

[0050] Figure 4 This is a schematic diagram of a vehicle lateral deviation in an embodiment of this disclosure, such as... Figure 4 As shown, e d e represents the first lateral deviation of the target vehicle's rear axle center point relative to the initial reference trajectory. ψ The heading deviation at the rear axle center point. L1 represents the second lateral deviation of the front axle center point relative to the initial reference trajectory. L1 represents the vehicle's wheelbase. This indicates the distance from the center of the rear axle to the rear edge of the vehicle. δ represents the distance from the center of the front axle to the leading edge of the vehicle, W represents the width of the vehicle, s represents the arc length along the initial reference trajectory from the projection point of the rear axle center point on the initial reference trajectory, and δ is the front wheel deflection angle of the target vehicle.

[0051] In this embodiment, the first lateral deviation, heading deviation, and second lateral deviation at the sampling point together constitute the state vector of that sampling point. Compared to describing the state vector of the target vehicle using only the first lateral deviation of the rear axle center point relative to the initial reference trajectory, describing the state vector using the lateral deviation of the rear axle center point and a preset position point is more suitable for the motion characteristics of large and medium-sized vehicles, better describes the deviation of the vehicle body relative to the initial reference trajectory, and thus ensures the safety of the planned trajectory.

[0052] The second lateral deviation can be determined based on the first lateral deviation and the heading deviation. Specifically, the first lateral deviation, the heading deviation, and the second lateral deviation have a circular arc transformation relationship; that is, the circular arc transformation relationship can be obtained by approximating it with a circular arc, and the second lateral deviation can be determined using the first lateral deviation, the heading deviation, and the circular arc transformation relationship. In this embodiment, the kinematic model can be a model describing the tracking deviation, and can include the circular arc transformation relationship. That is, the kinematic model can determine the second lateral deviation of each sampling point based on the circular arc transformation relationship, the first lateral deviation of the sampling point, and the heading deviation.

[0053] In addition to circular arc transformation relationships, kinematic models can also include state vector transformation relationships. These state vector transformation relationships can be used to determine the state vectors of adjacent sampling points based on the state vectors of the sampling points. For example, the state vector of the next sampling point can be determined based on the state vectors of the sampling points and the state vector transformation relationships.

[0054] For example, the state vector transition relationship can be represented by the following formula:

[0055]

[0056]

[0057]

[0058] in, Let represent the derivative of the variable with respect to time, i.e., the change in amount; s represents the arc length along the initial reference trajectory by which the projection of the rear axis center point onto the initial reference trajectory moves; e represents the arc length along the initial reference trajectory. y For the first lateral deviation, e ψ For heading deviation, v represents the velocity at the center of the rear axle of the vehicle, κ r (s) represents the reference curvature of the projection point of the rear axle center point onto the initial reference trajectory, which can be abbreviated as κ. r δ is the front wheel deflection angle of the vehicle, and the relationship between the actual steering curvature κ of the target vehicle and the front wheel deflection angle of the vehicle is κ=tanδ / L1.

[0059] Transforming the above formula to the Frenet coordinate system, we get:

[0060] e′ d =(1-e d k r )tane ψ ;

[0061]

[0062] Where, (·)′=d(·) / ds, represents the differential of the variable with respect to the arc length s, e d Indicates the first lateral deviation, e ψ Let κ represent the heading deviation and κ represent the actual steering curvature. Using the above formula, the rate of change of the first lateral deviation and the heading deviation relative to the arc length s can be obtained. It should be noted that the purpose of transforming the formula to the Frenet coordinate system is twofold: firstly, to eliminate the influence of the velocity factor; and secondly, to better describe the tracking deviation of the rear axle center point relative to the initial reference trajectory, i.e., the first lateral deviation.

[0063] Furthermore, the state vector transition relationship can also be expressed by the following formula:

[0064] χ i+1 =f(χ) i ,κ i );

[0065] Where, χ i Let χ be the state vector at the i-th sampling point. i+1 Let i be the state vector at the (i+1)th sampling point. κ i Let be the actual turning curvature at the i-th sampling point.

[0066] For example, for the above formula, we can linearize it by expanding it using Taylor and keeping only the first-order terms, resulting in:

[0067] χ i+1 =A i χ i +Bκ i ;

[0068] In the formula,

[0069] By constructing the kinematic model through the above-mentioned arc transformation relationship and state vector transformation relationship, a kinematic model that can be used to determine the second lateral deviation of each sampling point and the state vector of adjacent sampling points is obtained. Compared with the kinematic model that only describes the tracking deviation of the rear axle center point, the kinematic model of this embodiment can better describe the tracking deviation of large and medium-sized vehicles.

[0070] Optionally, the arc transformation relationship satisfies the following formula:

[0071]

[0072] Among them, c s =s+R ref sine ψ , Indicates the second lateral deviation; The coordinates of the center of the circle fitted to the initial reference trajectory. R is the actual arc length corresponding to the preset position point, that is, the actual arc length corresponding to the projection point of the preset position point on the initial reference trajectory. ref Let be the turning radius of the initial reference trajectory, s be the actual arc length corresponding to the rear axle center point, and e be the turning radius. d For the first lateral deviation, e ψ This is the heading deviation.

[0073] Specifically, the aforementioned circular arc transformation relationship can describe the expression for the second lateral deviation of the front axle center point relative to the initial reference trajectory in the Frenet coordinate system. In the Frenet coordinate system, the lateral deviation of each point on the target vehicle's wheelbase relative to the initial reference trajectory can be approximated by a circular arc. For example, see... Figure 5 , Figure 5 This is an approximate schematic diagram of tracking deviation in the Frenet coordinate system according to an embodiment of this disclosure, wherein, The coordinates of the center of the circle fitted to the initial reference trajectory. For the contour points on the target vehicle, This represents the actual arc length corresponding to each contour point.

[0074] By using the above formula for the arc transformation relationship, the second lateral deviation of the preset position point under each sampling point can be determined. The first lateral deviation, the second lateral deviation, and the heading deviation are used to measure the actual state of the target vehicle at each sampling point, so as to achieve an accurate description of the deviation of the vehicle body relative to the initial reference trajectory.

[0075] In this embodiment, the safety constraint can be a constraint to prevent the target vehicle from colliding with the road boundary or obstacles of the area to be planned.

[0076] In one specific implementation, before performing sequential secondary planning, the method further includes: constructing a safety constraint for the target vehicle based on the fact that the sum of a preset safety distance and the third lateral deviation of the contour point relative to the initial reference trajectory does not exceed a preset collision constraint; wherein the contour point is a position point on the inner or outer side of the target vehicle, and the preset collision constraint consists of road boundary constraints and obstacle constraints.

[0077] The preset safety distance can be a pre-set safety distance threshold. The third lateral deviation of the contour point relative to the initial reference trajectory can be the distance between the contour point and its projection point on the initial reference trajectory. The contour point can be any point on the inner or outer side of the vehicle.

[0078] Specifically, safety constraints can be expressed by the following formula: Where, d s Pχ represents the third lateral deviation, which is the preset safety distance. This represents the preset collision constraints consisting of road boundary constraints and obstacle constraints.

[0079] In the above safety constraints, the purpose of the safety constraints is to ensure that the third lateral deviation of the position point on the inner or outer side of the target vehicle relative to the initial reference trajectory satisfies the preset collision constraint - preset safety distance. That is, the third lateral deviation needs to be within the range of the road boundary constraint and obstacle constraint minus the preset safety distance.

[0080] By limiting the third lateral deviation to be within the range of road boundary constraints and obstacle constraints minus a preset safety distance, the vehicle body can be kept within the road boundary when planning the trajectory, and the distance between the vehicle body and the road boundary or obstacle meets a certain safety distance threshold, thereby avoiding collisions between the target vehicle and the road boundary or obstacle and improving the safety of the target vehicle passing through the planned area.

[0081] Optionally, the third lateral deviation is obtained based on the first lateral deviation and the heading deviation, and the calculation of the third lateral deviation satisfies the following formula:

[0082]

[0083] Among them, c s =s+R ref sine ψ , This represents the third lateral deviation of the m-th profile point. The coordinates of the center of the circle fitted to the initial reference trajectory. R is the actual arc length corresponding to the contour point. ref e is the turning radius of the initial reference trajectory. d For the first lateral deviation, e ψ For the heading deviation, r ± When the contour point is located on the inner side of the target vehicle, take R. ref -W / 2,r ± When the contour point is located on the outer side of the target vehicle, take R. ref +W / 2, where W is the width of the target vehicle.

[0084] Specifically, for contour points on the outer or inner side of the vehicle, the third lateral deviation of the contour points can be determined using circular arc fitting. Specifically, the third lateral deviation of contour points on the outer and inner sides of the vehicle can be calculated using the following formulas:

[0085]

[0086]

[0087] In the formula, The third lateral deviation of the contour points on the outer side of the vehicle body. This represents the third lateral deviation of the contour points on the inner side of the vehicle body. M represents the number of contour points on the inner and outer sides of the vehicle, where m∈(0,1,...,M-1).

[0088] The above formula enables the accurate determination of the third lateral deviation, which in turn facilitates the calculation of the difference between the third lateral deviation and the preset collision constraint and the preset safety distance based on the safety constraints.

[0089] Specifically, in this embodiment, the objective function is to maximize the uniformity of the projection distribution of the initial reference trajectory, and the kinematic model and safety constraints are used as limiting conditions. Sequential quadratic programming is then performed to obtain a sequence of target control quantities, including the actual steering curvature corresponding to each sampling point. The uniformity of the projection distribution can be determined by the area of ​​the target vehicle's projection onto the plane of the initial reference trajectory on both sides of the trajectory.

[0090] In one specific implementation, the objective function is to maximize the uniformity of the projection distribution of the initial reference trajectory, and the kinematic model of the target vehicle and safety constraints are used as limiting conditions. Sequential quadratic programming is then performed to obtain a sequence of target control variables, including:

[0091] Based on the weighted sum of the first and second lateral deviations of each sampling point, the safety margin reference value of each sampling point is determined. The sum of the safety margin reference values ​​of each sampling point is determined as the traffic safety level, which is used to describe the uniformity of the projected distribution. Taking the minimization of the traffic safety level of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence.

[0092] Specifically, the uniformity of the projected distribution can be described using a traffic safety factor, where the traffic safety factor is the sum of the safety margin reference values ​​for each sampling point. The safety margin reference values ​​are the weighted sum of the first lateral deviation and the second lateral deviation. The weights corresponding to the second lateral deviation and the second horizontal deviation can be set values, or they can be calculated based on the turning radius of the rear axle center point.

[0093] That is, the uniformity of the projection distribution can be described by the first lateral deviation and the second lateral deviation. The smaller the safety margin reference value of each sampling point, the higher the safety margin of each sampling point, that is, the higher the safety margin for the target vehicle to pass at each sampling point, and thus the lower the passage safety, the greater the uniformity of the projection distribution, and the higher the safety of the target vehicle traveling along the trajectory.

[0094] In the above implementation, the uniformity of the projection distribution is described by the traffic safety degree. Under the condition of satisfying the kinematic model and safety constraints, the optimal control quantity sequence that can minimize the traffic safety degree is obtained. There is no need to calculate the area of ​​the target vehicle distributed on both sides of the trajectory separately. The weighted sum of the first lateral deviation and the second lateral deviation can be calculated. While ensuring the trajectory safety, the trajectory planning efficiency is improved.

[0095] Optionally, a safety margin reference value for each sampling point is determined based on the weighted sum of the first lateral deviation and the second lateral deviation. This includes: determining the turning radius of the rear axle center point based on the distance from the preset position point to the leading edge of the target vehicle, the wheelbase of the target vehicle, and the width of the target vehicle; determining the weight corresponding to the second lateral deviation based on the turning radius of the rear axle center point, the turning radius of the initial reference trajectory, and the wheelbase of the target vehicle; determining the product of the weight corresponding to the second lateral deviation and the second lateral deviation; and determining the safety margin reference value based on the sum of the product and the first lateral deviation.

[0096] The turning radius of the rear axle center point is determined based on the distance from the preset position point to the leading edge of the target vehicle, the wheelbase of the target vehicle, and the width of the target vehicle, and can satisfy the following formula:

[0097]

[0098] Where R1 is the turning radius of the rear axle center point, L1 is the wheelbase of the target vehicle, and R ref W is the turning radius of the initial reference trajectory, and W is the width of the target vehicle.

[0099] Specifically, the formula for determining the turning radius of the rear axle center point can be derived using the following formula, see [link to formula]. Figure 6 , Figure 6 This is a schematic diagram of relevant parameters for a safety margin reference value in an embodiment of this disclosure. For any turning radius R1 at the rear axle center point, the following can be obtained based on geometric relationships:

[0100]

[0101]

[0102] Among them, R 1,r This refers to the maximum turning radius on the target vehicle, specifically the turning radius corresponding to the point on the target vehicle furthest from the initial reference trajectory. Figure 6 As shown, the upper left corner of the target vehicle corresponds to a turning radius of R. bus,r ;R 1,l This refers to the minimum turning radius on the target vehicle. Specifically, it could be the turning radius corresponding to the point on the target vehicle closest to the initial reference trajectory, such as... Figure 6 As shown, the lower right tire of the target vehicle corresponds to a turning radius of R. bus,l . Figure 6 R in road R1 and R2 are the turning radius corresponding to the initial reference trajectory and the turning radius corresponding to the rear axle center point, respectively. e is the second lateral deviation of the front axle center point. yThis is the first lateral deviation of the rear axle center point.

[0103] Furthermore, to ensure that the projection of the target vehicle onto the plane of the initial reference trajectory is distributed as evenly as possible on both sides of the initial reference trajectory, the turning radius of the initial reference trajectory must satisfy the following:

[0104]

[0105] Combining the above formulas, we can obtain the formula for determining the turning radius of the rear axle center point. Furthermore, the weight corresponding to the second lateral deviation can be calculated based on the turning radius of the rear axle center point, the turning radius of the initial reference trajectory, and the wheelbase. For example, see the following formula:

[0106]

[0107] Among them, G k This represents the weight corresponding to the second lateral deviation. Further, the safety margin reference value... Through the above method, the accurate calculation of the safety margin reference value at each sampling point was achieved, and thus the optimization solution with the minimization of the sum of the safety margin reference values ​​as the objective function was realized.

[0108] It should be noted that the process of sequential quadratic programming can be an iterative optimization solution process. Sequential quadratic programming can be methods such as SQP (Sequential Quadratic Programming), IPOPT (InteriorPoint OPTimizer), or CiLQR (Constrained Iterative Linear Quadratic Regulator).

[0109] Optionally, taking the minimization of the traffic safety of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as limitations, sequential quadratic programming is performed to obtain the target control quantity sequence, including:

[0110] The objective function is to minimize the traffic safety of the initial reference trajectory, and the kinematic model and safety constraints of the target vehicle are used as constraints to determine the current control quantity sequence. It is then determined whether the iteration cutoff condition is met. If not, the initial reference trajectory is updated based on the current control quantity sequence, and the process returns to the previous steps of minimizing traffic safety, determining the current control quantity sequence, and using the kinematic model and safety constraints as constraints, until the iteration cutoff condition is met and the target control quantity sequence is obtained.

[0111] The iteration cutoff condition can be that the number of iterations reaches a preset number, or that the difference in the passability of the initial reference trajectory obtained after each iteration gradually converges.

[0112] Specifically, the lane lines can be used as the initial reference trajectory. Then, the current control quantity sequence can be solved based on the objective function and constraints. Based on the current control quantity sequence and the kinematic model, the initial reference trajectory can be redefined. Then, the objective function can be maximized again using the projection distribution uniformity of the newly defined initial reference trajectory, with the kinematic model and safety constraints as constraints. The current control quantity sequence can be solved again. If the iteration cutoff condition is not met, the initial reference trajectory can be determined again based on the current control quantity sequence and the kinematic model. The above process is repeated until the iteration cutoff condition is met. The last determined current control quantity sequence is taken as the target control quantity sequence.

[0113] It should be noted that when updating the initial reference trajectory, the sampling points in the initial reference trajectory can also be updated simultaneously. The actual steering curvature corresponding to each sampling point in the target control quantity sequence can be the actual steering curvature issued before the target vehicle's projection point reaches the next sampling point. Through the above iterative optimization solution process, the optimal control quantity sequence can be obtained after multiple iterations, thereby improving the safety of the trajectory.

[0114] In addition to considering the safety of the trajectory, this embodiment can also consider the smoothness of the trajectory, so as to plan a trajectory with high smoothness while ensuring safety.

[0115] In one specific implementation, the method further includes: determining a ride comfort reference value for a sampling point based on the difference between the actual steering curvature of the sampling point and the actual steering curvature of the previous sampling point; determining the ride comfort as the sum of the ride comfort reference values ​​of each sampling point; determining a ride reference quantity based on the ride comfort and ride safety; taking the minimization of the ride reference quantity of the initial reference trajectory as the objective function; and using the kinematic model of the target vehicle and safety constraints as the limiting conditions, performing sequential quadratic programming to obtain a sequence of target control quantities.

[0116] Specifically, the square of the difference between the actual steering curvature of a sampling point and the previous sampling point can be used as the smoothness reference value for that sampling point. The sum of the average smoothness reference values ​​for all sampling points is then used as the driving smoothness. In essence, the lower the driving smoothness, the higher the trajectory smoothness.

[0117] Furthermore, the sum of traffic smoothness and traffic safety is used as the traffic reference value. Minimizing the traffic reference value is the objective function. Under the constraint conditions, the target control quantity sequence is solved. Through this implementation method, the optimal control quantity sequence with both low traffic smoothness and low traffic safety can be obtained, thereby planning a trajectory with high safety and smoothness.

[0118] Considering that excessively large or varying actual steering curvature can affect the smoothness and safety of the target vehicle traversing the planned area, the method provided in this embodiment may optionally include: using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle, safety constraints, a first steering curvature constraint, and a second steering curvature constraint as limiting conditions, performing sequential quadratic programming to obtain a sequence of target control quantities; wherein, the first steering curvature constraint is used to constrain the actual steering curvature of each sampling point from not exceeding a preset threshold, and the second steering curvature constraint is used to constrain the difference between the actual steering curvatures of adjacent sampling points from not exceeding a preset variation.

[0119] That is, in addition to the kinematic model and safety constraints, the constraints may also include a first steering curvature constraint and a second steering curvature constraint. The first steering curvature constraint is used to ensure that the actual steering curvature does not exceed a preset threshold, such as |κ... i |≤κ max ,i∈(0,1,...,N-1). The second steering curvature constraint means that the actual change in steering curvature does not exceed the preset change, such as |κ i -κ i-1 |≤Δκ max ,i∈(0,1,...,N-1), where N is the number of sampling points.

[0120] In the above implementation, by using the first steering curvature constraint and the second steering curvature constraint as constraints, the optimal control quantity sequence that satisfies the first steering curvature constraint and the second steering curvature constraint can be solved, further improving the safety and smoothness of the planned trajectory.

[0121] S130. Determine the target reference trajectory based on the target control quantity sequence, and determine each tracking point that minimizes the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory. The tracking points are the position points on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same.

[0122] Specifically, after obtaining the target control quantity sequence, the target reference trajectory with the same turning curvature as the actual turning curvature at each sampling point can be obtained through each actual turning curvature in the target control quantity sequence.

[0123] Furthermore, by using each actual steering curvature, the tracking point for achieving each actual steering curvature can be obtained. This tracking point can be a position used to control the vehicle's movement along the target reference trajectory.

[0124] It should be noted that, in this embodiment, speed control information during the travel of the target reference trajectory can also be determined. Specifically, speed and acceleration information can be used together as a state vector, and then, after solving the target control quantity sequence, speed and acceleration information at each sampling point can be obtained through the kinematic model and the target control quantity sequence.

[0125] After obtaining the target reference trajectory and each tracking point in the target reference trajectory, the target reference trajectory and each tracking point can be sent to the autonomous driving control system of the target vehicle, so that the autonomous driving control system can control the target vehicle to drive along the target reference trajectory through each tracking point.

[0126] The vehicle trajectory planning method provided in this embodiment obtains the initial reference trajectory of the target vehicle in the area to be planned. The objective function is to maximize the average distribution of the target vehicle's projection on the plane of the initial reference trajectory on both sides of the trajectory. Using a kinematic model and safety constraints as limitations, sequential secondary planning is performed to obtain a sequence of target control quantities, including the actual steering curvature corresponding to each sampling point. This sequence of target control quantities is then used to determine the target reference trajectory and the tracking points that are not entirely identical during the driving process along the target reference trajectory. The goal is to obtain the target reference trajectory and the tracking points that minimize the difference in distance between the two sides of the vehicle relative to the trajectory during the driving process. This avoids the problem that when the target vehicle is traveling along the target reference trajectory, the distance on one side of the vehicle relative to the trajectory is too large, leading to a collision risk due to that side being too close to the road boundary or roadside obstacle in the area. This ensures the vehicle can safely pass through narrow areas and solves the problem in the prior art where it is difficult to plan a trajectory for long-wheelbase vehicles to safely pass through narrow areas.

[0127] Figure 7 This is a schematic diagram of the structure of a vehicle trajectory planning device according to an embodiment of this disclosure. Figure 7 As shown: The device includes: an initial trajectory determination module 710, a sequential secondary planning module 720, and a target trajectory determination module 730.

[0128] The initial trajectory determination module 710 is used to obtain the initial reference trajectory of the target vehicle in the area to be planned, wherein the initial reference trajectory includes each sampling point;

[0129] The sequential quadratic programming module 720 is used to perform sequential quadratic programming with the objective function of maximizing the uniformity of the projection distribution of the initial reference trajectory and the kinematic model and safety constraints of the target vehicle as the limiting conditions, to obtain a target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point;

[0130] The target trajectory determination module 730 is used to determine a target reference trajectory based on the target control quantity sequence, and to determine each tracking point that minimizes the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory, wherein the tracking point is a position point on the target vehicle traveling along the target reference trajectory, and the tracking points are not completely the same.

[0131] The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances on both sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point.

[0132] The vehicle trajectory planning device provided in this embodiment can execute the steps in the vehicle trajectory planning method provided in this embodiment, and has the execution steps and beneficial effects, which will not be repeated here.

[0133] Figure 8 This is a schematic diagram of the structure of an electronic device according to an embodiment of this disclosure. See below for details. Figure 8 It shows a schematic diagram of a structure suitable for implementing the electronic device 500 in the embodiments of this disclosure. Figure 8 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0134] like Figure 8 As shown, the electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 501, which can perform various appropriate actions and processes to implement the methods of the embodiments described herein, based on a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the electronic device 500. The processing device 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0135] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts, thereby implementing the vehicle trajectory planning method as described above. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by the processing device 501, it performs the functions defined in the methods of embodiments of this disclosure.

[0136] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, 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 device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0137] The aforementioned computer-readable 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 medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0138] Obtain the initial reference trajectory of the target vehicle within the area to be planned, wherein the initial reference trajectory includes each sampling point;

[0139] Using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point;

[0140] Based on the target control quantity sequence, a target reference trajectory is determined, and each tracking point is determined to minimize the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory. The tracking points are the position points on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same.

[0141] The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances on both sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point.

[0142] Optionally, when one or more of the above-described procedures are executed by the electronic device, the electronic device may also perform other steps described in the above embodiments.

[0143] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on 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 of the foregoing.

[0144] Option 1: A vehicle trajectory planning method, the method comprising:

[0145] Obtain the initial reference trajectory of the target vehicle within the area to be planned, wherein the initial reference trajectory includes each sampling point;

[0146] Using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point;

[0147] Based on the target control quantity sequence, a target reference trajectory is determined, and each tracking point is determined to minimize the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory. The tracking points are the position points on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same.

[0148] The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances on both sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point.

[0149] Option 2: According to the method described in Option 1, before performing sequential secondary planning, the method further includes:

[0150] Determine the circular arc transformation relationship between the first lateral deviation of the rear axle center point of the target vehicle relative to the initial reference trajectory, the heading deviation of the rear axle center point, and the second lateral deviation of the preset position point on the target vehicle relative to the initial reference trajectory;

[0151] The first lateral deviation, the heading deviation, and the second lateral deviation are determined as state vectors, and the state vector transformation relationship between the state vectors of adjacent sampling points is determined.

[0152] Based on the arc transformation relationship and the state vector transformation relationship, the kinematic model of the target vehicle is constructed.

[0153] Option 3: According to the method described in Option 2, the arc transformation relationship satisfies the following formula:

[0154]

[0155] Among them, c s =s+R ref sine ψ , This indicates the second lateral deviation; The coordinates of the center of the circle fitted to the initial reference trajectory. R is the actual arc length corresponding to the preset position point. ref Let be the turning radius of the initial reference trajectory, s be the actual arc length corresponding to the rear axle center point, and e be the turning radius of the initial reference trajectory. d e is the first lateral deviation. ψ The heading deviation is mentioned above.

[0156] Option 4: Based on the method described in Option 2, using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as limitations, sequential quadratic programming is performed to obtain the target control quantity sequence, including:

[0157] Based on the weighted sum of the first lateral deviation and the second lateral deviation of each sampling point, a safety margin reference value for each sampling point is determined, and the sum of the safety margin reference values ​​of each sampling point is determined as the passage safety level, so as to describe the uniformity of the projected distribution through the passage safety level.

[0158] Using the minimization of the traffic safety of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence.

[0159] Option 5: According to the method described in Option 4, the step of minimizing the traffic safety of the initial reference trajectory as the objective function, and using the kinematic model of the target vehicle and safety constraints as limitations, performing sequential quadratic programming to obtain the target control quantity sequence includes:

[0160] The current control sequence is determined by taking the minimization of the traffic safety of the initial reference trajectory as the objective function and the kinematic model and safety constraints of the target vehicle as the limiting conditions.

[0161] Determine whether the iteration cutoff condition is met. If not, update the initial reference trajectory based on the current control quantity sequence, and return to execute the steps of determining the current control quantity sequence with the objective function of minimizing the passage safety, the kinematic model and the safety constraints as the limiting conditions, until the iteration cutoff condition is met and the target control quantity sequence is obtained.

[0162] Option 6: According to the method described in Option 4, determining the safety margin reference value for each sampling point based on the weighted sum of the first lateral deviation and the second lateral deviation of each sampling point includes:

[0163] The turning radius of the rear axle center point is determined based on the distance from the preset position point to the front edge of the target vehicle, the wheelbase of the target vehicle, and the width of the target vehicle.

[0164] The weight corresponding to the second lateral deviation is determined based on the turning radius of the rear axle center point, the turning radius of the initial reference trajectory, and the wheelbase of the target vehicle.

[0165] Determine the product of the weight corresponding to the second lateral deviation and the second lateral deviation, and determine the safety margin reference value based on the sum of the product and the first lateral deviation.

[0166] Option 7: According to the method described in Option 2, before performing sequential secondary planning, the method further includes;

[0167] The safety constraints of the target vehicle are constructed based on the fact that the sum of the preset safety distance and the third lateral deviation of the contour point relative to the initial reference trajectory does not exceed the preset collision constraint.

[0168] The contour points are the locations on the inner or outer side of the target vehicle, and the preset collision constraints consist of road boundary constraints and obstacle constraints.

[0169] Option 8: According to the method described in Option 7, the third lateral deviation is obtained based on the first lateral deviation and the heading deviation, and the calculation of the third lateral deviation satisfies the following formula:

[0170]

[0171] Among them, c s =s+R ref sine ψ , This represents the third lateral deviation of the m-th profile point. The coordinates of the center of the circle fitted to the initial reference trajectory. R is the actual arc length corresponding to the contour point. ref e is the turning radius of the initial reference trajectory. d e is the first lateral deviation. ψ Let r be the heading deviation. ± When the contour point is located on the inner side of the target vehicle, R is taken. ref -W / 2,r ± When the contour point is located on the outer side of the target vehicle, take R. ref +W / 2, where W is the width of the target vehicle.

[0172] Option 9: The method described in Option 4, further comprising:

[0173] Based on the difference between the actual steering curvature of the sampling point and the actual steering curvature of the previous sampling point, the ride comfort reference value of the sampling point is determined, and the sum of the ride comfort reference values ​​of all the sampling points is determined as the driving ride comfort.

[0174] Based on the smoothness of traffic and the safety of traffic, a traffic reference quantity is determined. The objective function is to minimize the traffic reference quantity of the initial reference trajectory. The kinematic model of the target vehicle and the safety constraints are used as limiting conditions. Sequential quadratic programming is performed to obtain the target control quantity sequence.

[0175] Option 10: The method according to Option 1, further comprising:

[0176] Using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model, safety constraints, first steering curvature constraints, and second steering curvature constraints of the target vehicle as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence.

[0177] Wherein, the first steering curvature constraint is used to constrain the actual steering curvature of each of the sampling points from not exceeding a preset threshold, and the second steering curvature constraint is used to constrain the difference between the actual steering curvatures of adjacent sampling points from not exceeding a preset change amount.

[0178] Option 11: A vehicle trajectory planning device, comprising:

[0179] An initial trajectory determination module is used to obtain an initial reference trajectory of the target vehicle in the area to be planned, wherein the initial reference trajectory includes each sampling point;

[0180] The sequential quadratic programming module is used to perform sequential quadratic programming with the objective function of maximizing the uniformity of the projection distribution of the initial reference trajectory and the kinematic model and safety constraints of the target vehicle as the limiting conditions, to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point;

[0181] The target trajectory determination module is used to determine a target reference trajectory based on the target control quantity sequence, and to determine each tracking point that minimizes the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory, wherein the tracking point is a position point on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same;

[0182] The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances on both sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point.

[0183] Option 12: An electronic device, the electronic device comprising:

[0184] One or more processors;

[0185] Storage device for storing one or more programs;

[0186] When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of schemes 1-10.

[0187] Option 13: A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of Options 1-10.

[0188] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A vehicle trajectory planning method, characterized in that, The method includes: Obtain the initial reference trajectory of the target vehicle within the area to be planned, wherein the initial reference trajectory includes each sampling point; Using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point; Based on the target control quantity sequence, a target reference trajectory is determined, and each tracking point is determined to minimize the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory. The tracking points are the position points on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same. The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances between the two sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point. Before performing sequential quadratic programming, the method further includes: Determine the circular arc transformation relationship between the first lateral deviation of the rear axle center point of the target vehicle relative to the initial reference trajectory, the heading deviation of the rear axle center point, and the second lateral deviation of the preset position point on the target vehicle relative to the initial reference trajectory; The first lateral deviation, the heading deviation, and the second lateral deviation are determined as state vectors, and the state vector transformation relationship between the state vectors of adjacent sampling points is determined. Based on the arc transformation relationship and the state vector transformation relationship, the kinematic model of the target vehicle is constructed.

2. The method according to claim 1, characterized in that, The arc transformation relationship satisfies the following formula: ； in, , ; This indicates the second lateral deviation; The coordinates of the center of the circle fitted to the initial reference trajectory. The actual arc length corresponding to the preset position point. Let be the turning radius of the initial reference trajectory. The actual arc length corresponding to the center point of the rear axle. This is the first lateral deviation. The heading deviation is mentioned above.

3. The method according to claim 1, characterized in that, Using the maximization of the uniformity of the projection distribution of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as limitations, sequential quadratic programming is performed to obtain the target control variable sequence, including: Based on the weighted sum of the first lateral deviation and the second lateral deviation of each sampling point, a safety margin reference value for each sampling point is determined, and the sum of the safety margin reference values ​​of each sampling point is determined as the passage safety level, so as to describe the uniformity of the projected distribution through the passage safety level. Using the minimization of the traffic safety of the initial reference trajectory as the objective function, and the kinematic model of the target vehicle and safety constraints as the limiting conditions, sequential quadratic programming is performed to obtain the target control quantity sequence.

4. The method according to claim 3, characterized in that, The objective function is to minimize the traffic safety of the initial reference trajectory, and the kinematic model and safety constraints of the target vehicle are used as limiting conditions. Sequential quadratic programming is then performed to obtain the target control quantity sequence, including: The current control sequence is determined by minimizing the traffic safety of the initial reference trajectory as the objective function and by using the kinematic model of the target vehicle and safety constraints as the limiting conditions. Determine whether the iteration cutoff condition is met. If not, update the initial reference trajectory based on the current control quantity sequence, and return to execute the steps of determining the current control quantity sequence with the objective function of minimizing the passage safety, the kinematic model and the safety constraints as the limiting conditions, until the iteration cutoff condition is met and the target control quantity sequence is obtained.

5. The method according to claim 3, characterized in that, The step of determining the safety margin reference value for each sampling point based on the weighted sum of the first lateral deviation and the second lateral deviation of each sampling point includes: The turning radius of the rear axle center point is determined based on the distance from the preset position point to the front edge of the target vehicle, the wheelbase of the target vehicle, and the width of the target vehicle. The weight corresponding to the second lateral deviation is determined based on the turning radius of the rear axle center point, the turning radius of the initial reference trajectory, and the wheelbase of the target vehicle. Determine the product of the weight corresponding to the second lateral deviation and the second lateral deviation, and determine the safety margin reference value based on the sum of the product and the first lateral deviation.

6. The method according to claim 1, characterized in that, Before performing sequential quadratic programming, the method further includes: The safety constraints of the target vehicle are constructed based on the fact that the sum of the preset safety distance and the third lateral deviation of the contour point relative to the initial reference trajectory does not exceed the preset collision constraint. The contour points are the locations on the inner or outer side of the target vehicle, and the preset collision constraints consist of road boundary constraints and obstacle constraints.

7. A vehicle trajectory planning device, characterized in that, include: An initial trajectory determination module is used to obtain an initial reference trajectory of the target vehicle in the area to be planned, wherein the initial reference trajectory includes each sampling point; The sequential quadratic programming module is used to perform sequential quadratic programming with the objective function of maximizing the uniformity of the projection distribution of the initial reference trajectory and the kinematic model and safety constraints of the target vehicle as the limiting conditions, to obtain the target control quantity sequence, wherein the target control quantity sequence includes the actual steering curvature corresponding to each sampling point; The target trajectory determination module is used to determine a target reference trajectory based on the target control quantity sequence, and to determine each tracking point that minimizes the difference between the distances on both sides of the vehicle relative to the target reference trajectory during the driving process of the target reference trajectory, wherein the tracking point is a position point on the target vehicle traveling along the target reference trajectory, and the tracking points are not exactly the same; The projection distribution uniformity is the average degree to which the projection of the target vehicle on the plane of the initial reference trajectory is distributed on both sides of the trajectory. It is used to reflect the difference between the distances between the two sides of the vehicle relative to the initial reference trajectory. The projection distribution uniformity is determined based on the state vector of each sampling point. The state vector of the sampling point is determined based on the kinematic model, the state vector of the adjacent sampling point, and the actual steering curvature of the adjacent sampling point. The construction of the kinematic model includes: Determine the circular arc transformation relationship between the first lateral deviation of the rear axle center point of the target vehicle relative to the initial reference trajectory, the heading deviation of the rear axle center point, and the second lateral deviation of the preset position point on the target vehicle relative to the initial reference trajectory; The first lateral deviation, the heading deviation, and the second lateral deviation are determined as state vectors, and the state vector transformation relationship between the state vectors of adjacent sampling points is determined. Based on the arc transformation relationship and the state vector transformation relationship, the kinematic model of the target vehicle is constructed.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images