A path planning method, device and computer storage medium

Abstract

The application relates to the computer technical field, in particular to a path planning method and device and computer storage medium; the method comprises the following steps: performing motion modeling based on the position of a target rear wheel and the maximum deflection angle of a target front wheel to obtain the motion constraint of a moving object; obtaining the space constraint of the moving object based on a planning terminal point and space information around the moving object; performing information fusion on the motion constraint and the space constraint to obtain a fusion constraint; determining the respective running parameters of a plurality of feasible space points based on the fusion constraint; performing path planning based on the respective running parameters of the plurality of feasible space points to obtain a target path between the starting position of the target rear wheel and the planning terminal point; the obstacles around the moving object and the motion constraint of the moving object are fused into one constraint, and the path planning is performed based on the fusion constraint, so that the path planning efficiency is improved, and the accuracy and executability of the planning path are improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a path planning method, apparatus and computer storage medium. Background Technology

[0002] Path planning is an essential functional module for autonomous driving of vehicles or autonomous driving of mobile robots. In existing technologies, traditional graph search methods or analytical optimal curve methods are commonly used for path planning. For example, the A* algorithm or Dijkstra's algorithm in traditional graph search methods usually only considers stationary obstacles around the moving object, without considering the motion constraints of the moving object itself. Although the Reeds-Sheep curve or Dubins curve in analytical optimal curve methods considers the motion constraints of the moving object itself, they do not consider the obstacles around the moving object. As a result, the planned path may not conform to the actual driving or motion scenario. Summary of the Invention

[0003] To address the aforementioned problems in the prior art, the purpose of this application is to integrate the obstacles around the moving object and the motion constraints of the moving object itself into a single constraint, and to perform path planning based on the integrated constraint, thereby improving the efficiency of path planning and enhancing the accuracy and feasibility of the planned path.

[0004] To address the aforementioned problems, this application provides a path planning method applied to a moving object including target front wheel steering, the method comprising:

[0005] Motion modeling is performed based on the position of the target's rear wheel and the maximum deflection angle of the target's front wheel to obtain the motion constraints of the moving object;

[0006] Based on the planning endpoint and the spatial information around the moving object, spatial constraints on the moving object are obtained; the spatial constraints characterize that the moving object can move without obstruction in the feasible state space; the feasible state space includes multiple feasible spatial points.

[0007] The motion constraints and the spatial constraints are fused to obtain fused constraints.

[0008] Based on the fusion constraints, the operating parameters corresponding to each of the multiple feasible spatial points are determined.

[0009] Path planning is performed based on the operating parameters corresponding to each of the multiple feasible spatial points to obtain the target path from the starting position of the target rear wheel to the planned endpoint.

[0010] In this embodiment of the application, obtaining the spatial constraints on the moving object based on the planned endpoint and the spatial information surrounding the moving object includes:

[0011] Based on the spatial information surrounding the moving object, the moving object is modeled and analyzed to determine the feasible state space of the moving object;

[0012] Based on the planned endpoint and the starting position of the target rear wheel, the path of the moving object is simulated to determine the target driving trajectory corresponding to the moving object;

[0013] Based on the feasible state space and the target driving trajectory, the spatial constraints of the moving object are determined.

[0014] In this embodiment of the application, the information fusion of the motion constraints and the spatial constraints to obtain fused constraints includes:

[0015] Based on the spatial constraints of the moving object, dynamic planning is performed on the movement process of the moving object to obtain the relationship between the spatial constraints and the motion constraints;

[0016] The motion constraints and the associated relationships are fused to obtain the fused constraints.

[0017] In this embodiment of the application, determining the operating parameters corresponding to each of the plurality of feasible spatial points based on the fusion constraints includes:

[0018] Based on the preset spatial resolution and the discretization of the feasible state space, the plurality of feasible spatial points are determined;

[0019] Based on the fusion constraints, constraint analysis is performed on the multiple feasible space points to obtain the operating parameters corresponding to each of the multiple feasible space points.

[0020] In this embodiment of the application, the operating parameters include steering control variables, and the path planning based on the operating parameters corresponding to the plurality of feasible spatial points to obtain the target path from the starting position of the target rear wheel to the planned endpoint includes:

[0021] Based on the steering control amount corresponding to the starting position of the target rear wheel and the motion constraints, path planning is performed to obtain the current intermediate point;

[0022] Determine the current path based on the starting position and the current intermediate point;

[0023] Based on the steering control amount of the current intermediate point and the motion constraints, path planning is performed to obtain the next intermediate point;

[0024] Update the current path based on the next intermediate point;

[0025] The next intermediate point is determined as the current intermediate point;

[0026] Repeated execution: Based on the steering control amount of the current intermediate point and the motion constraints, perform path planning to obtain the next intermediate point, until the next intermediate point is determined as the current intermediate point, until the current intermediate point is the planning endpoint;

[0027] The current path is determined to be the target path.

[0028] In this embodiment of the application, the operating parameters include the minimum travel time and the steering control quantity corresponding to the minimum travel time. The step of performing constraint analysis on the plurality of feasible space points based on the fusion constraints to obtain the operating parameters corresponding to each of the plurality of feasible space points includes:

[0029] Based on the fusion constraints, constraint analysis is performed on the multiple feasible spatial points to determine the travel time of the moving object from any feasible spatial point to the planned destination; the travel time includes the time corresponding to each of the multiple movement modes.

[0030] Based on the time corresponding to each of the various modes of movement, determine the minimum travel time corresponding to each of the multiple feasible spatial points;

[0031] Based on the minimum travel time corresponding to each of the multiple feasible spatial points, determine the steering control quantity corresponding to the minimum travel time;

[0032] The operating parameters corresponding to each of the multiple feasible spatial points are determined based on the minimum travel time corresponding to each of the multiple feasible spatial points and the steering control quantity corresponding to the minimum travel time.

[0033] In this embodiment of the application, the step of solving the constraints on the plurality of feasible spatial points based on the fusion constraints to determine the travel time of the moving object from any feasible spatial point to the planned destination includes:

[0034] Determine the initial analysis values ​​corresponding to each of the plurality of feasible spatial points;

[0035] Based on the constraint fusion and the initial analysis values ​​corresponding to the feasible points adjacent to the target feasible point, the constraints of the target feasible point are solved to obtain the updated analysis values ​​corresponding to the target feasible point; the target feasible point can be any feasible point.

[0036] The updated analysis value is determined as the initial analysis value;

[0037] Repeated execution: Based on the constraint fusion and the initial analysis values ​​corresponding to the adjacent feasible points of the target feasible point, the constraint solution is performed on the target feasible point to obtain the analysis update value corresponding to the target feasible point; the analysis update value is determined as the initial analysis value; until the number of constraint solutions is greater than the preset number of solutions, or the analysis error corresponding to each of the multiple feasible points is less than the preset error;

[0038] The initial analysis value corresponding to each of the multiple feasible spatial points is determined as the travel time of the moving object from any feasible spatial point to the planned destination.

[0039] On the other hand, this application also provides a path planning device, the device comprising:

[0040] The motion constraint construction module is used to perform motion modeling based on the position of the target rear wheel and the maximum deflection angle of the target front wheel, so as to obtain the motion constraints of the moving object;

[0041] A spatial constraint construction module is used to obtain spatial constraints on the moving object based on the planning endpoint and the spatial information around the moving object; the spatial constraints characterize that the moving object can move without obstruction in the feasible state space; the feasible state space includes multiple spatial points;

[0042] The constraint information fusion module is used to fuse the motion constraints and the spatial constraints to obtain fused constraints.

[0043] The operation parameter determination module is used to determine the operation parameters corresponding to each of the plurality of feasible spatial points based on the fusion constraints.

[0044] The path planning module is used to perform path planning based on the operating parameters corresponding to each of the multiple feasible spatial points, so as to obtain the target path from the starting position of the target rear wheel to the planned endpoint.

[0045] On the other hand, this application also provides an electronic device, the device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the path planning method as described above.

[0046] On the other hand, this application also provides a computer storage medium storing at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded and executed by a processor to implement the path planning method described above.

[0047] Due to the above technical solutions, the path planning method described in this application has the following beneficial effects:

[0048] By constructing motion constraints for the moving object based on its front and rear wheels, and fusing these constraints with spatial constraints determined by the surrounding spatial information, a unique constraint for the moving object is obtained. This fusion of motion and spatial constraints into a single constraint avoids the need to process multiple constraints in path planning, thus improving efficiency. Furthermore, by determining the operational parameters corresponding to multiple feasible spatial points through the fusion constraint, path planning is performed based on these parameters, making the target path more consistent with actual driving scenarios and improving the accuracy and feasibility of the target path planning. Attached Figure Description

[0049] To more clearly illustrate the technical solutions of this application, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0050] Figure 1 This is a schematic diagram of a path planning method provided in an embodiment of this application;

[0051] Figure 2 This is a schematic diagram of the angle between the moving object and the coordinate axis in a path planning method provided in this application embodiment;

[0052] Figure 3 This is a flowchart illustrating the determination of the feasible state space in a path planning method provided in an embodiment of this application.

[0053] Figure 4 This is a schematic diagram of the process for determining fusion constraints in a path planning method provided in an embodiment of this application;

[0054] Figure 5 This is a flowchart illustrating the determination of operating parameters in a path planning method provided in an embodiment of this application;

[0055] Figure 6 This is a flowchart illustrating the determination of a target path in a path planning method provided in an embodiment of this application;

[0056] Figure 7 This is a flowchart illustrating the determination of operating parameters in a path planning method provided in an embodiment of this application;

[0057] Figure 8 This is a flowchart illustrating the determination of travel time in a route planning method provided in an embodiment of this application;

[0058] Figure 9This is a result diagram of a path planning method provided in this application being applied to an automatic parking scenario;

[0059] Figure 10 This is a schematic diagram of a path planning device provided in an embodiment of this application;

[0060] Figure 11 This is a hardware structure block diagram of a path planning method provided in an embodiment of this application. Detailed Implementation

[0061] 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 some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0062] The term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of this application. In the description of this application, it should be understood that the terms "upper," "lower," "left," "right," "top," "bottom," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, 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 indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. Moreover, the terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein.

[0063] Combination Figure 1 This application introduces a path planning method provided by an embodiment, applicable to a moving object including a target front wheel and a target rear wheel. Specifically, the moving object requires coordinated movement of both the target front and rear wheels; the target rear wheel has a small rotation amplitude, while the target front wheel has a large rotation amplitude, meaning the method is applied to a moving object where the target front wheel is used for steering. For example, the moving object can be a car, tricycle, bicycle, or a robot with front and rear wheels. The method includes:

[0064] S1001. Based on the position of the target's rear wheel and the maximum deflection angle of the target's front wheel, motion modeling is performed to obtain the motion constraints of the moving object.

[0065] In this embodiment, the position of the target rear wheel can refer to the location of the center point of the rear wheel; in the case of multiple target rear wheels, it can be the center of the line connecting the center points of the multiple target rear wheels, the location of any one of the center points of the target rear wheel, or other positions of the target rear wheel, which are not limited here; the maximum deflection angle of the target front wheel refers to the maximum rotation angle δ of the target front wheel relative to the direction of movement of the moving object. max .

[0066] In a specific embodiment of this application, S1001 includes: determining the minimum turning radius of the moving object based on the distance between the centers of the front and rear wheels and the maximum deflection angle of the target front wheel; performing motion modeling based on the center position of the target rear wheel and the minimum turning radius of the moving object to obtain the motion constraints of the moving object.

[0067] Specifically, refer to the following formula (1) to calculate the minimum turning radius of the moving object based on the distance between the centers of the front and rear wheels and the maximum deflection angle of the target front wheel:

[0068] R = L / tanδ max (1)

[0069] Where R refers to the minimum turning radius of the moving object, L refers to the distance between the centers of the front and rear wheels of the moving object, and δ max It refers to the maximum deflection angle of the target's front wheels. In a vehicle, the maximum deflection angle of the target's front wheels is directly proportional to the maximum steering wheel angle.

[0070] refer to Figure 2 In this embodiment of the application, if a rectangular coordinate system is constructed based on any direction, then the center position of the target rear wheel is the coordinate axis (x, y) of the aforementioned position in the rectangular coordinate system; the angle between the moving object and the coordinate x-axis is θ, that is, the angle between the line connecting the target front wheel and the target rear wheel and the x-axis is θ, and the deflection angle of the target front wheel relative to the moving object is δ, which is determined by the steering control amount of the moving object.

[0071] In a specific embodiment of this application, referring to formula (2), based on the position of the target rear wheel and the minimum turning radius of the moving object, motion modeling is performed to obtain the motion constraints of the moving object, namely the motion constraint formula (2):

[0072]

[0073] Where (x, y) refers to the position of the target's rear wheel, θ refers to the angle between the line connecting the target's front wheel and the target's rear wheel and the x-axis, and (x, y, θ) represents the state vector of the moving object; The derivative of the state vector of the moving object with respect to time is denoted by v; v is the linear velocity of the moving object per unit motion, with a value range of (-1∪1); s is the steering control variable of the moving object, with a value range of [-1,1]; R is the minimum turning radius of the moving object.

[0074] In a specific embodiment of this application, when s = 0, it represents that the moving object is moving in a straight line.

[0075] S1002. Based on the planning endpoint and the spatial information around the moving object, spatial constraints on the moving object are obtained; spatial constraints represent that the moving object can move without hindrance in the feasible state space; the feasible state space includes multiple feasible spatial points.

[0076] In this embodiment, the planned endpoint can be one or more pre-input pose points, which is the state vector (x, y, θ) corresponding to the endpoint of the path planning of the moving object; the spatial information around the moving object refers to stationary obstacles, road conditions, road information, etc. around the moving object, specifically vehicles, garage edges, walls, and the thickness of water / ice on the ground; in specific embodiments, the spatial information around the moving object can be acquired by the moving object's own sensors or obtained from other information platforms, which is not limited here; the moving object can move safely and without obstruction in the feasible state space, specifically including but not limited to not colliding with obstacles such as vehicles and walls, and not being unable to drive normally due to poor road conditions (deep water, thick ice).

[0077] In this embodiment of the application, if the feasible state space is placed in the constructed rectangular coordinate system, multiple feasible space points and the spatial coordinates corresponding to the multiple feasible space points can be obtained.

[0078] S1003. Information fusion is performed on motion constraints and spatial constraints to obtain fused constraints.

[0079] In this embodiment of the application, information fusion refers to the fusion of motion constraints and spatial constraints into fusion constraints, which can characterize the constraints on the movement of a moving object.

[0080] S1004. Determine the operating parameters corresponding to each of the multiple feasible spatial points based on the fusion constraints.

[0081] In this embodiment of the application, the operating parameters refer to the parameters generated during the process of the moving object moving to a feasible spatial point. The operating parameters include, but are not limited to, travel time, steering control amount, running speed and direction.

[0082] S1005. Based on the operating parameters corresponding to multiple feasible spatial points, perform path planning to obtain the target path from the starting position of the target rear wheel to the planned endpoint.

[0083] In this embodiment, path planning refers to determining the desired trajectory of a moving object. Specifically, it can be planning the shortest target path for the moving object. The starting position of the target's rear wheel and the starting steering angle of the target's front wheel can be obtained based on the moving object's own sensors or derived from the collected data of the moving object itself. No limitation is made here. The target path can be an optimal driving path or multiple driving paths, depending on the number of planned endpoints and the path planning requirements. For example, the path planning requirement is the shortest path and the minimum energy consumption.

[0084] In this embodiment, the motion constraints of the moving object are constructed based on the front and rear wheels of the moving object, and the motion constraints of the moving object are fused with the spatial constraints determined based on the spatial information around the moving object to obtain the unique constraints of the moving object. The motion constraints and spatial constraints are fused into a fused constraint, thereby avoiding the processing of multiple constraints in path planning and improving the efficiency of path planning. By determining the corresponding operating parameters of multiple feasible spatial points through the fused constraint, and performing path planning based on the corresponding operating parameters of multiple feasible spatial points, the target path is made more consistent with the actual driving scenario, thereby improving the accuracy and executability of the target path planning.

[0085] refer to Figure 3 In this embodiment of the application, S1002 includes:

[0086] S3001. Based on the spatial information around the moving object, model and analyze the moving object to determine the feasible state space of the moving object.

[0087] In this embodiment of the application, modeling and analyzing the moving object refers to simulating the process of the moving object moving from its starting position to any position in the surrounding space according to the motion constraint formula (2). Specifically, the modeling and analysis includes spatial and moving object modeling, collision detection analysis, ground adhesion coefficient analysis, etc. Finally, the space around the moving object is divided into feasible state space and infeasible state space. The moving object can move safely and without hindrance in the feasible state space. The infeasible state space refers to the space in which the moving object may overlap or collide with other obstacles during the movement, and may stall due to ground factors, thus affecting normal driving.

[0088] S3002. Based on the planned endpoint and the starting position of the target rear wheel, perform path simulation on the moving object to determine the target driving trajectory corresponding to the moving object.

[0089] In this embodiment of the application, path simulation of a moving object refers to the process of simulating the movement of the moving object from its starting position to the planned destination in the feasible state space, and the target driving trajectory includes multiple simulated trajectories.

[0090] S3003. Based on the feasible state space and the target driving trajectory, determine the spatial constraints of the moving object.

[0091] In this embodiment of the application, the spatial constraint of the moving object can specifically be that the moving object can move without collision in the feasible state space, and the motion of the moving object satisfies the motion constraint formula (2).

[0092] In a specific embodiment of this application, the optimal motion cost determined based on the feasible state space and motion constraints can be characterized by the following spatial constraint formula (3); the optimal motion cost characterizes the optimal motion parameters of the moving object under spatial constraints, such as the shortest motion time, the shortest motion distance, and the lowest motion energy consumption.

[0093] V(x)=min{T:X(t)∈Ω(X),t∈[0,T],X(T)∈Xfree} (3)

[0094] Where V(x) represents the minimum time taken for a moving object to travel along a trajectory that starts from state X and satisfies both motion and spatial constraints; t represents the time it takes for the moving object to travel to the planned endpoint based on the target trajectory; x(t) represents the moving trajectory of the moving object, which can be in vector form; Ω(x) represents the set of target trajectories corresponding to the moving object; T represents the maximum time it takes for the moving object to travel to the planned endpoint based on the target trajectory; Xfree represents the feasible state space; X(T)∈Xfree indicates that the moving trajectory of the moving object is located in the feasible state space.

[0095] In a specific embodiment of this application, V(x) is the optimal value function corresponding to the heuristic function in path planning.

[0096] In this embodiment, it is assumed that the moving object travels at a unit linear velocity. Therefore, the minimum travel time of the moving object is equivalent to the shortest travel trajectory.

[0097] In this embodiment, the spatial constraints of the moving object are determined based on the feasible state space and the target driving trajectory, thereby making the spatial constraints of the moving object more concrete and improving the reliability of path planning. Preferably, the spatial constraints of the moving object are represented by the value function of the heuristic function to further realize the specific calculation of path planning, thereby improving the reliability and practicality of path planning.

[0098] refer to Figure 4In this embodiment of the application, S1003 includes:

[0099] S4001. Based on the spatial constraints of the moving object, perform dynamic programming on the moving process of the moving object to obtain the relationship between spatial constraints and motion constraints.

[0100] In this embodiment of the application, the specific process of dynamic programming for the movement of the moving object can be found in the following formula (4). Based on the principle of dynamic programming, the recursive form of V(X) can be obtained:

[0101] V(X)=min(V(X(Δt))+Δt) (4)

[0102] Here, V(X) represents the minimum time taken for a moving object to complete its trajectory, starting from state X and satisfying both motion and spatial constraints.

[0103] Based on basic operational logic, formula (4) can be transformed to obtain formula (5):

[0104]

[0105] When Δt approaches 0, that is, when Δt→0, we can obtain the following formula (6):

[0106]

[0107] in, The gradient vector of the characterization function V(X) with respect to the state vector X = (x, y, θ) of the moving object; The derivative of the moving state vector X with respect to time, specifically,

[0108] In a specific embodiment of this application, formula (6) represents the relationship between spatial constraints and motion constraints.

[0109] S4002. Information fusion is performed on motion constraints and relationships to obtain fused constraints.

[0110] In this embodiment of the application, the information fusion process refers to the process of fusing and calculating motion constraints and related relationships to obtain fused constraints that include motion constraints and spatial constraints. Specifically, the motion constraint formula (2) and... Substituting into formula (6), we can obtain the fusion constraint formula (7):

[0111] min{V x v cosθ+V y v sinθ+V θ s|v| / R+1}=0 (7)

[0112] Where v refers to the unit linear velocity of the moving object, and its value ranges from (-1∪1); s refers to the steering control quantity of the moving object, and its value ranges from [-1,1].

[0113] In this embodiment, the fusion constraint formula (7) is the HJB function (Hamilton-Jacobi-Bellman, heuristic function) in the path planning process.

[0114] In this embodiment, by generating fused constraints including motion constraints and spatial constraints, the constraints in path planning are made more consistent with the actual situation, which not only reduces the complexity of path planning but also improves its reliability. Specifically, the motion heuristic function represents the fused constraints, further realizing the specific calculation of path planning, thereby improving the reliability and practicality of path planning.

[0115] In a specific embodiment of this application, the HJB function does not have a continuously differentiable solution everywhere, therefore its weak solution should be sought. To this end, the min symbol is removed from formula (7), resulting in the equivalent fusion constraint formula (8), which is used for subsequent numerical solutions to the weak solution (viscous solution) of the HJB function:

[0116] -|V x cosθ+V y sinθ|-|V θ | / R-1=0 (8)

[0117] refer to Figure 5 In this embodiment of the application, S1004 includes:

[0118] S5001. Based on the preset spatial resolution and the feasible state space, discretize to determine multiple feasible space points.

[0119] In this embodiment of the application, a preset spatial resolution (h) is used. x h y h θ A three-dimensional mesh is constructed in the feasible state space to achieve the discretization of the feasible state space, resulting in multiple discrete feasible space points. Preferably, the multiple discrete feasible space points are uniformly discretized on the three-dimensional mesh to obtain multiple spatial grid points. The preset spatial resolution can be (0.01, 0.01, 0.01) or (0.02, 0.02, 0.02), and the specific value is determined based on actual needs and is not limited here.

[0120] In this embodiment of the application, V i，j，k =(i*h x j*h y ,k*h θ) represents the discrete value of the value function at the (i,j,k)th grid point; preferably, V i，j，k An approximation of the value function representing the value at the (i,j,k)th grid point.

[0121] S5002. Based on fusion constraints, constraint analysis is performed on multiple feasible space points to obtain the corresponding operating parameters for each of the multiple feasible space points.

[0122] In this embodiment of the application, constraint analysis of multiple feasible space points refers to obtaining the running parameters corresponding to each of the multiple feasible space points when the moving object moves to them under the constraints of the fused constraints.

[0123] In this embodiment of the application, the feasible state space is discretized, thereby facilitating subsequent numerical calculations for path planning and improving the reliability and practicality of path planning.

[0124] In a specific embodiment of this application, the specific formula for discretization based on the fusion constraint formula (8) is as follows:

[0125] When the vehicle is traveling in the forward direction, i.e., v = 1:

[0126]

[0127]

[0128] Where, n1 = sign(cos(kh) θ )), n2=sign(sin(kh θ )).

[0129] Specifically, sign(x) is the sign function. When x>0, sign(x)=1; when x=0, sign(x)=0; when x<0, sign(x)=-1.

[0130] When the vehicle is reversing, i.e., when v = -1:

[0131]

[0132]

[0133] For the third dimension θ, |V θ The discretized expression formula for | is as follows:

[0134]

[0135] In a specific embodiment of this application, the gradient vector of the value function with respect to the state vector of the moving object is represented by two adjacent spatial grid points, thereby making the target path more consistent with the actual motion scenario.

[0136] refer to Figure 6 In this embodiment of the application, the operating parameters include steering control quantity, and S1005 includes:

[0137] S6001. Based on the steering control quantity corresponding to the starting position of the target rear wheel and the motion constraints, perform path planning to obtain the current intermediate point.

[0138] In this embodiment of the application, path planning refers to determining the current intermediate point adjacent to the starting point position based on the steering control amount of the starting point of the path planning under motion constraints.

[0139] In a specific embodiment of this application, the operating parameters also include the unit motion linear velocity. Specifically, the unit motion linear velocity and the steering control amount corresponding to the starting position of the target rear wheel are substituted into the motion constraint formula (1), and the motion constraint formula (1) is integrated to obtain (x, y, θ), that is, the current intermediate point (x, y) is obtained.

[0140] S6002. Determine the current path based on the starting position and the current intermediate point;

[0141] In this embodiment of the application, the current path is the line connecting the starting position and the current intermediate point.

[0142] S6003. Based on the steering control amount of the current intermediate point and the motion constraints, perform path planning to obtain the next intermediate point.

[0143] In this embodiment of the application, the path planning process is the process of inputting the steering control quantity into the motion constraint formula and performing numerical integration.

[0144] S6004. Update the current path based on the next intermediate point.

[0145] S6005. Set the next intermediate point as the current intermediate point.

[0146] Repeat steps S6003-S6005 until the current intermediate point is the planned endpoint.

[0147] S6006. Determine the current path as the target path.

[0148] In this embodiment of the application, if the position coordinates of the current intermediate point are equal to the coordinates of the planned endpoint, it indicates that all points on the target path have been determined, and thus the current path can be determined as the target path.

[0149] In this embodiment, the current intermediate point is obtained by path planning based on the steering control amount corresponding to the starting position of the target rear wheel and motion constraints; the next intermediate point is obtained by path planning based on the steering control amount corresponding to the current intermediate point and motion constraints. In other words, the path planning in this application is to progressively plan multiple feasible spatial points in the feasible state space under motion constraints, thereby making the target path more consistent with the actual driving scenario, improving the accuracy of the planned path, and increasing the reliability of the target path; at the same time, progressively planning the target path improves the accuracy of the target path.

[0150] In this embodiment of the application, the operating parameters also include the minimum travel time and the steering control quantity corresponding to the minimum travel time. It should be noted that the steering control quantity corresponding to the minimum travel time is the steering control quantity required to be calculated in S6001-S6006. Specifically, the minimum travel time corresponding to any feasible spatial point represents the minimum time for any feasible spatial point to travel to the planned endpoint.

[0151] refer to Figure 7 In this embodiment of the application, S5002 includes:

[0152] S7001. Based on fusion constraints, perform constraint analysis on multiple feasible spatial points to determine the travel time of a moving object from any feasible spatial point to the planned destination; the travel time includes the time corresponding to each of the various modes of movement.

[0153] In this embodiment, the travel time of a moving object from any feasible spatial point to the planned endpoint is related to the movement mode of the moving object. For example, the travel time corresponding to a vehicle traveling in a straight line is different from that corresponding to a vehicle traveling in a curved line; the travel time corresponding to a vehicle traveling in a straight line is different from that corresponding to a vehicle traveling in reverse in a straight line. Therefore, multiple movement modes include: traveling in a straight line, traveling in a curved line, traveling in reverse in a straight line, and traveling in reverse in a curved line. It should be noted that traveling in a curved line refers to the change of the angle θ between the connecting line between the target front wheel and the target rear wheel of the moving object and the x-axis.

[0154] In a specific embodiment of this application, during forward straight-line driving, the linear velocity of the moving object is v = 1, and the rate of change of the included angle θ is 0; during forward curved driving, the unit linear velocity of the moving object is v = 1, and the rate of change of the included angle θ is not 0; during reverse straight-line driving, the unit linear velocity of the moving object is v = -1, and the rate of change of the included angle θ is 0; during reverse curved driving, the unit linear velocity of the moving object is v = -1, and the rate of change of the included angle θ is not 0.

[0155] In a specific embodiment of this application, constraint analysis of multiple feasible spatial points refers to solving V based on multiple movement methods. i,j,kValue functions; see formulas (13)-(16) below for details:

[0156] When v = 1, |V θ When | = 0:

[0157]

[0158] When v = 1, |V θ When |≠0:

[0159]

[0160] When v = -1, |V θ When |=0:

[0161]

[0162] When v = -1, |V θ When |≠0:

[0163]

[0164] Where, a=|cos(kh) θ )| / h x b = |sin(kh) θ )| / h y c = 1 / (h) θ R).

[0165] Then the travel time V of the moving object from any feasible spatial point to the planned destination is... i,j,k = (V1, V2, V3, V4).

[0166] S7002. Based on the time corresponding to each of the various modes of movement, determine the minimum travel time corresponding to each of the multiple feasible spatial points.

[0167] S7003. Based on the minimum travel time corresponding to each of the multiple feasible spatial points, determine the steering control quantity corresponding to the minimum travel time.

[0168] S7004. Based on the minimum travel time corresponding to each of the multiple feasible spatial points and the steering control quantity corresponding to the minimum travel time, determine the operating parameters corresponding to each of the multiple feasible spatial points.

[0169] In this embodiment of the application, by performing constraint analysis on the various movement modes of the moving object, and determining the minimum travel time corresponding to each of the multiple feasible spatial points based on the time corresponding to each of the various movement modes, it is possible to fully consider the travel time of the moving object to the planned destination under various operating conditions, and select the case with the shortest travel time, thereby facilitating the selection of the route with the shortest path to the planned destination and improving the accuracy of the target path.

[0170] refer to Figure 8 In this embodiment of the application, S7001 includes:

[0171] S8001. Determine the initial values ​​for analysis corresponding to each of the multiple feasible spatial points.

[0172] In this embodiment, the initial analysis value represents the assumed time to reach the planning endpoint. Specifically, the initial analysis value at the planning endpoint is 0, and the further away from the planning endpoint, the larger the initial analysis value. In this specific embodiment, except for the initial analysis value at the planning endpoint, the initial analysis value can be greater than 10. 6 It can also be greater than 10 3 .

[0173] S8002. Based on constraint fusion and the initial analysis values ​​corresponding to adjacent feasible points in the target feasible space, the constraint solution is performed on the target feasible space point to obtain the updated analysis values ​​corresponding to the target feasible space point; the target feasible space point is selected as any feasible space point.

[0174] In a specific embodiment of this application, the formula for calculating the updated value is as follows:

[0175] V i,j,k =min(V1, V2, V3, V4, V i,j,k’ (17)

[0176] Among them, V i,j,k To analyze the updated values; V i,j,k’ The initial values ​​for analysis, i.e. the values ​​obtained in the previous iteration, (V1, V2, V3, V4) represent the travel time of the moving object from any feasible spatial point to the planned endpoint under the initial values ​​for analysis.

[0177] In this embodiment of the application, the linear velocity of motion is determined by comparing the updated analysis value with the initial analysis value. Specifically, when min(V1,V2)... <min(V3,V4,V ijk When v = 1, otherwise v = -1. The formula for calculating the steering control amount corresponding to the minimum driving time is as follows: s = sign(V i,j,k ).

[0178] S8003. Determine the analysis update value as the analysis initial value.

[0179] In this embodiment of the application, constraining the target feasible space points is achieved by using formulas (13)-(16) to solve for the travel time V among the feasible space points. i,j,k Iterative updates will be performed.

[0180] Repeat steps S8002-S8003 until the number of constraint solutions exceeds the preset number of solutions, or the analysis errors corresponding to each of the multiple feasible space points are less than the preset error.

[0181] In this embodiment, the preset number of solutions is equivalent to the number of constraint solution loops. Specifically, the number of constraint solution loops can be 100 or 1000, and the specific value is not limited here. The preset error can be 0.1 or 0.2, and the specific value is not limited here. The analysis error corresponding to each of the multiple feasible space points is determined based on the current initial analysis value and the initial analysis value of the previous loop. Specifically, the formula for calculating the analysis error is:

[0182] e = (V i,j,k -V i,j,k’ (17)

[0183] Where e is the analysis error, V i,j,k To analyze the updated values; V i,j,k’ To analyze the initial values.

[0184] S8004. Determine the initial analysis values ​​corresponding to each of the multiple feasible spatial points, which represent the travel time of the moving object from any feasible spatial point to the planned endpoint.

[0185] In this embodiment, by setting initial analysis values ​​and performing rapid iterative scanning on these initial values, and when the analysis errors are all less than the preset error or the number of constraint solutions is greater than the preset number of solutions, the travel time of the moving object from any feasible spatial point to the planned endpoint is determined. This improves the accuracy of the calculation of the travel time of the moving object from any feasible spatial point to the planned endpoint, thereby improving the feasibility of the target path and increasing the efficiency of path planning.

[0186] refer to Figure 9 In a specific embodiment of this application, taking the automatic parking scenario of a vehicle as an example, the path planning method in this application can accurately obtain the shortest target path for reversing into the parking space; the path planning method in this application improves the reliability of path planning for moving objects and improves the efficiency of path planning.

[0187] refer to Figure 10 This application also provides a path planning device, which is applied to a moving object including a target front wheel and a target rear wheel. The device includes:

[0188] The motion constraint construction module 101 is used to perform motion modeling based on the position of the target rear wheel and the maximum deflection angle of the target front wheel, so as to obtain the motion constraints of the moving object.

[0189] The spatial constraint construction module 102 is used to obtain spatial constraints on the moving object based on the planning endpoint and the spatial information around the moving object; the spatial constraint represents that the moving object can move without hindrance in the feasible state space; the feasible state space includes multiple feasible spatial points;

[0190] The constraint information fusion module 103 is used to fuse motion constraints and spatial constraints to obtain fused constraints;

[0191] The running parameter determination module 104 is used to determine the running parameters corresponding to each of the multiple feasible spatial points based on the fusion constraints.

[0192] The path planning module 105 is used to perform path planning based on the operating parameters corresponding to multiple feasible spatial points, so as to obtain the target path from the starting position of the target rear wheel to the planned endpoint.

[0193] The spatial constraint building module includes:

[0194] The simulation operation unit is used to model and analyze the moving object based on the spatial information around the moving object, and to determine the feasible state space of the moving object;

[0195] The path simulation unit is used to simulate the path of a moving object based on the planned endpoint and the starting position of the target rear wheel, and to determine the target driving trajectory corresponding to the moving object.

[0196] The spatial constraint determination unit is used to determine the spatial constraints of a moving object based on the feasible state space and the target driving trajectory.

[0197] The constraint information fusion module includes:

[0198] The dynamic programming unit is used to perform dynamic programming on the movement process of a moving object based on its spatial constraints, and to obtain the relationship between spatial constraints and motion constraints.

[0199] The information fusion unit is used to fuse motion constraints and relationships to obtain fused constraints.

[0200] The operating parameter determination module includes:

[0201] The spatial mapping unit is used to discretize the feasible state space based on a preset spatial resolution and to determine multiple feasible spatial points.

[0202] The constraint analysis unit is used to perform constraint analysis on multiple feasible space points based on fused constraints, and obtain the corresponding operating parameters for each of the multiple feasible space points.

[0203] Operating parameters include steering control parameters, and the path planning module includes:

[0204] The first path planning unit is used to plan the path based on the steering control quantity corresponding to the starting position of the target rear wheel and the motion constraints, and obtain the current intermediate point.

[0205] The current path determination unit is used to determine the current path based on the starting position and the current intermediate point.

[0206] The second path planning unit is used to perform path planning based on the steering control quantity of the current intermediate point and motion constraints to obtain the next intermediate point;

[0207] The current path update unit is used to update the current path based on the next intermediate point;

[0208] The current intermediate point determination unit is used to determine the next intermediate point as the current intermediate point;

[0209] The first loop unit is used to repeatedly execute: based on the steering control quantity of the current intermediate point and the motion constraints, perform path planning to obtain the next intermediate point, until the next intermediate point is determined as the current intermediate point, until the current intermediate point is the planning endpoint;

[0210] The target path determination unit is used to determine the current path as the target path.

[0211] The operating parameters include the minimum travel time and the corresponding steering control quantity. The constraint analysis unit includes:

[0212] The travel time determination unit is used to perform constraint analysis on multiple feasible spatial points based on fusion constraints, and determine the travel time of the moving object from any feasible spatial point to the planned destination; the travel time includes the time corresponding to each of the various modes of movement;

[0213] The minimum travel time determination unit is used to determine the minimum travel time for each of the multiple feasible spatial points based on the time corresponding to each of the various modes of transportation.

[0214] The steering control quantity determination unit is used to determine the steering control quantity corresponding to the minimum travel time based on the minimum travel time corresponding to each of the multiple feasible spatial points.

[0215] The operating parameter determination unit is used to determine the operating parameters corresponding to each of the multiple feasible spatial points based on the minimum travel time corresponding to each of the multiple feasible spatial points and the steering control quantity corresponding to the minimum travel time.

[0216] The travel time determination unit includes:

[0217] The first analysis initial value determination unit is used to determine the analysis initial values ​​corresponding to each of the multiple feasible space points;

[0218] The constraint solving unit is used to solve the constraints of the target feasible space point based on constraint fusion and the analysis initial values ​​corresponding to the feasible space points adjacent to the target feasible space point, so as to obtain the analysis update value corresponding to the target feasible space point; the target feasible space point can be any feasible space point;

[0219] The second analysis initial value determination unit is used to determine the analysis update value as the analysis initial value;

[0220] The second loop module is used to repeatedly execute: based on constraint fusion and the initial analysis values ​​corresponding to adjacent feasible points in the target feasible space, the constraint solution is performed on the target feasible space point to obtain the analysis update value corresponding to the target feasible space point; the analysis update value is determined as the analysis initial value; until the number of constraint solutions is greater than the preset number of solutions, or the analysis error corresponding to each of the multiple feasible space points is less than the preset error;

[0221] The target travel time determination unit is used to determine the initial analysis values ​​corresponding to each of the multiple feasible spatial points, which are the travel times for the moving object from any feasible spatial point to the planned destination.

[0222] This application also provides an electronic device, which includes a processor and a memory. The memory stores at least one instruction or at least one program, which is loaded and executed by the processor to implement the path planning method described above.

[0223] Memory is used to store software programs and modules. The processor executes these stored software programs and modules to perform various functional applications and data processing. Memory can primarily consist of a program storage area and a data storage area. The program storage area stores the operating system, application programs required for functionality, etc.; the data storage area stores data created based on device usage, etc. Furthermore, memory can include high-speed random access memory (RAM) and non-volatile memory, such as at least one hard disk drive, flash memory, or other volatile solid-state storage devices. Correspondingly, memory can also include a memory controller to provide the processor with access to the memory.

[0224] The methods and embodiments provided in this application can be executed in electronic devices such as mobile terminals, computer terminals, servers, or similar computing devices. Figure 11 This is the electronic device provided in the embodiments of this application. For example... Figure 11As shown, the electronic device 900 can vary significantly due to differences in configuration or performance. It may include one or more central processing units (CPUs) 910 (CPUs 910 may include, but are not limited to, microprocessors such as MCUs or programmable logic devices such as FPGAs), a memory 930 for storing data, and one or more storage media 920 (e.g., one or more mass storage devices) for storing application programs 923 or data 922. The memory 930 and storage media 920 may be temporary or persistent storage. The program stored in the storage media 920 may include one or more modules, each module may include a series of instruction operations on the electronic device. Furthermore, the CPU 910 may be configured to communicate with the storage media 920 and execute the series of instruction operations in the storage media 920 on the electronic device 900. Electronic device 900 may also include one or more power supplies 960, one or more wired or wireless network interfaces 950, one or more input / output interfaces 940, and / or one or more operating systems 921, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0225] The input / output interface 940 can be used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the electronic device 900. In one example, the input / output interface 940 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the input / output interface 940 may be a radio frequency (RF) module for wireless communication with the Internet.

[0226] Those skilled in the art will understand that Figure 11 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device described above. For example, the electronic device 900 may also include... Figure 11 The more or fewer components shown, or having the same Figure 11 The different configurations shown.

[0227] Embodiments of this application also provide a storage medium storing at least one instruction or at least one program, wherein the at least one instruction or at least one program is loaded and executed by a processor to implement the path planning method as described above.

[0228] The foregoing description has fully disclosed the specific embodiments of this application. It should be noted that any modifications made by those skilled in the art to the specific embodiments of this application do not depart from the scope of the claims. Accordingly, the scope of the claims of this application is not limited to the foregoing specific embodiments.

Claims

1. A path planning method characterized by, The method is applied to a moving object including a target front wheel and a target rear wheel, and comprises the following steps: motion modeling is performed based on a position of the target rear wheel and a maximum deflection angle of the target front wheel to obtain a motion constraint of the moving object; a spatial constraint of the moving object is obtained based on a planning terminal point and spatial information around the moving object; the spatial constraint represents that the moving object can move without obstruction in a feasible state space; the feasible state space includes a plurality of feasible space points; information fusion is performed on the motion constraint and the spatial constraint to obtain a fused constraint; running parameters corresponding to the plurality of feasible space points are determined based on the fused constraint; the running parameters include a minimum travel time and a steering control amount corresponding to the minimum travel time; a target path between a starting position of the target rear wheel and the planning terminal point is planned based on the running parameters corresponding to the plurality of feasible space points; the determination of the running parameters corresponding to the plurality of feasible space points based on the fused constraint comprises the following steps: a plurality of feasible space points are determined based on a preset spatial resolution and the feasible state space; constraint analysis is performed on the plurality of feasible space points based on the fused constraint to determine travel times of the moving object from any feasible space point to the planning terminal point; the travel times include times corresponding to a plurality of moving modes; minimum travel times corresponding to the plurality of feasible space points are determined based on the times corresponding to the plurality of moving modes; a steering control amount corresponding to the minimum travel time is determined based on the minimum travel times corresponding to the plurality of feasible space points; the running parameters corresponding to the plurality of feasible space points are determined based on the minimum travel times corresponding to the plurality of feasible space points and the steering control amount corresponding to the minimum travel time.

2. The path planning method of claim 1, wherein, the obtaining of the spatial constraint of the moving object based on the planning terminal point and the spatial information around the moving object comprises the following steps: a feasible state space of the moving object is determined by modeling analysis of the moving object based on the spatial information around the moving object; a target travel trajectory corresponding to the moving object is determined by path simulation of the moving object based on the planning terminal point and a starting position of the target rear wheel; the spatial constraint of the moving object is determined based on the feasible state space and the target travel trajectory.

3. The path planning method of claim 2, wherein, the information fusion of the motion constraint and the spatial constraint to obtain the fused constraint comprises the following steps: an associated relationship between the spatial constraint and the motion constraint is determined by dynamic planning of a moving process of the moving object based on the spatial constraint of the moving object; the information fusion of the motion constraint and the associated relationship to obtain the fused constraint.

4. The path planning method of claim 1, wherein, the path planning based on the running parameters corresponding to the plurality of feasible space points to obtain the target path between the starting position of the target rear wheel and the planning terminal point comprises the following steps: a current intermediate point is obtained by path planning based on a steering control amount corresponding to the starting position of the target rear wheel and the motion constraint. determine a current path based on the start position and the current intermediate point; perform path planning based on a steering control amount of the current intermediate point and the motion constraint to obtain a next intermediate point; update the current path based on the next intermediate point; determine the next intermediate point as the current intermediate point; repeat the following until the current intermediate point is the planning end point: perform path planning based on a steering control amount of the current intermediate point and the motion constraint to obtain a next intermediate point, and determine the next intermediate point as the current intermediate point; determine the current path as the target path.

5. The path planning method of claim 1, wherein, The determining the travel time of the mobile object from any feasible space point to the planning end point based on the constraint fusion includes: determining an analysis initial value corresponding to each of the plurality of feasible space points; performing constraint solving on a target feasible space point based on the constraint fusion and an analysis initial value corresponding to a feasible space point adjacent to the target feasible space point to obtain an analysis update value corresponding to the target feasible space point; the target feasible space point is any feasible space point; determining the analysis update value as the analysis initial value; repeating the following until the constraint solving times are greater than a preset solving times or the analysis errors corresponding to each of the plurality of feasible space points are all less than a preset error: performing constraint solving on a target feasible space point based on the constraint fusion and an analysis initial value corresponding to a feasible space point adjacent to the target feasible space point to obtain an analysis update value corresponding to the target feasible space point, and determining the analysis update value as the analysis initial value; determining the analysis initial value corresponding to each of the plurality of feasible space points as the travel time of the mobile object from any feasible space point to the planning end point.

6. A route planning apparatus characterized by comprising: The device is applied to a mobile object including a target front wheel and a target rear wheel, and includes: a motion constraint construction module configured to perform motion modeling based on a position of the target rear wheel and a maximum deflection angle of the target front wheel to obtain a motion constraint of the mobile object; a space constraint construction module configured to obtain a space constraint of the mobile object based on a planning end point and space information around the mobile object; the space constraint represents that the mobile object can move in a feasible state space without obstruction; the feasible state space includes a plurality of feasible space points; a constraint information fusion module configured to perform information fusion on the motion constraint and the space constraint to obtain a constraint fusion; a running parameter determination module configured to determine a running parameter corresponding to each of the plurality of feasible space points based on the constraint fusion; the running parameter includes a minimum travel time and a steering control amount corresponding to the minimum travel time; a path planning module configured to perform path planning based on the running parameter corresponding to each of the plurality of feasible space points to obtain a target path between a start position of the target rear wheel and the planning end point; The determining the running parameter corresponding to each of the plurality of feasible space points based on the constraint fusion includes: determining the plurality of feasible space points based on a preset space resolution and discretization of the feasible state space; Conduct constraint analysis on the multiple feasible space points based on the fusion constraint, and determine driving time of the mobile object from any feasible space point to the planning terminal point; the driving time includes time corresponding to each of the multiple movement modes; Determine minimum driving time corresponding to each of the multiple feasible space points based on the time corresponding to each of the multiple movement modes; Determine steering control amount corresponding to the minimum driving time based on the minimum driving time corresponding to each of the multiple feasible space points; Determine running parameter corresponding to each of the multiple feasible space points based on the minimum driving time corresponding to each of the multiple feasible space points and the steering control amount corresponding to the minimum driving time.

7. A computer storage medium, characterized in that The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to implement the path planning method according to any one of claims 1-5.

8. An electronic device, comprising: The device includes a processor and a memory, and the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the path planning method according to any one of claims 1-5.

Images