Automatic driving vehicle path planning method and system based on electronic and electrical information architecture

Abstract

The invention discloses an automatic driving vehicle path planning method and system based on electronic and electrical information architecture, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining an initial vehicle state of an automatic driving vehicle and obstacle information in a driving environment; establishing a path planning model according to the initial vehicle state of the autonomous vehicle and the obstacle information in the driving environment; based on the path planning model and the electronic and electrical architecture model, performing global path planning on the autonomous vehicle by adopting an improved A * algorithm; based on the global path planning and the latest dynamic information of the external environment of the automatic driving vehicle, adopting a heuristic artificial potential field method to carry out local dynamic planning of the automatic driving vehicle; and performing path tracking control on the autonomous vehicle according to the global path planning and the local dynamic planning of the autonomous vehicle. According to the method provided by the invention, the optimality, safety and smoothness of planning are considered while the real-time performance is ensured.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method and system for autonomous driving vehicle path planning based on an electronic and electrical information architecture. Background Art

[0002] In the operation of the autonomous driving system, the design level of the planning layer determines the driving performance. First, a specific planning algorithm is used to generate a global path from the starting point to the target point based on the driving task requirements, scene map, and external environment information. In view of the dynamic driving environment, it is sometimes necessary to update the local path. Then, the autonomous driving vehicle drives automatically according to the path generated by the planning layer.

[0003] In the study of path planning methods, graph search algorithm is a representative method. Traditional graph search methods are suitable for global path solving in static environments. However, it is difficult to handle real-time changes in the local environment, and the path solved is often a series of discrete points. In order to solve the path planning problem in continuous space, the artificial potential field algorithm is used to generate real-time motion sequences. However, in a complex external environment, the path planned by this method often falls into the local optimum and cannot reach the target position. Although some variants of the artificial potential field method have been proposed to overcome the above problems, the impact on path smoothness and the additional computational burden still need to be considered. In summary, based on the characteristics of the above algorithms, it is of great value to propose a path planning that takes into account optimality, real-time and continuity. Summary of the invention

[0004] The purpose of this application is to provide a method and system for autonomous driving vehicle path planning based on electronic and electrical information architecture, which can ensure real-time performance while taking into account the optimality, safety and smoothness of planning.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for autonomous driving vehicle path planning based on an electronic and electrical information architecture, comprising:

[0007] Obtaining the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment;

[0008] A path planning model is established according to the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; the path planning model is composed of factors including the number of path points of the autonomous driving vehicle, the area occupied by the path points, the drivable area and the impassable area;

[0009] Based on the path planning model and the electronic and electrical architecture model, an improved A* algorithm is used to perform global path planning for the autonomous driving vehicle; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network connection system, an autonomous driving domain controller, a switch and a vehicle chassis domain controller; the perception system is used to perceive the dynamic information of the external environment of the autonomous driving vehicle; the positioning system is used to receive the positioning information of the autonomous driving vehicle and the global scene map; the intelligent network connection system is used to receive information in the entire traffic network; the autonomous driving domain controller is used to execute the intelligent driving function of the autonomous driving vehicle; the switch is used to complete data interaction with different domain controllers; the vehicle chassis domain controller is used to execute chassis operation control of the autonomous driving vehicle;

[0010] Based on the global path planning and the latest dynamic information of the external environment of the autonomous driving vehicle, a heuristic artificial potential field method is used to perform local dynamic planning of the autonomous driving vehicle;

[0011] The autonomous driving vehicle is controlled by path tracking according to the global path planning and local dynamic planning of the autonomous driving vehicle.

[0012] In a second aspect, the present application provides an autonomous driving vehicle path planning system based on an electronic and electrical information architecture, comprising:

[0013] A data acquisition module, used to obtain the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment;

[0014] A path planning model building module, used to establish a path planning model according to the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; the path planning model is composed of the number of path points of the autonomous driving vehicle, the area occupied by the path points, the drivable area and the impassable area;

[0015] A global path planning construction module is used to perform global path planning for the autonomous driving vehicle based on the path planning model and the electronic and electrical architecture model using an improved A* algorithm; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network connection system, an autonomous driving domain controller, a switch, and a vehicle chassis domain controller; the perception system is used to perceive dynamic information of the external environment of the autonomous driving vehicle; the positioning system is used to receive the positioning information of the autonomous driving vehicle and the global scene map; the intelligent network connection system is used to receive information from the entire traffic network; the autonomous driving domain controller is used to execute the intelligent driving function of the autonomous driving vehicle; the switch is used to complete data interaction with different domain controllers; the vehicle chassis domain controller is used to execute chassis operation control of the autonomous driving vehicle;

[0016] A local dynamic planning module, for performing local dynamic planning of the autonomous driving vehicle using a heuristic artificial potential field method based on global path planning and the latest dynamic information of the external environment of the autonomous driving vehicle;

[0017] A path tracking module is used to perform path tracking control on the autonomous driving vehicle based on the global path planning and local dynamic planning of the autonomous driving vehicle.

[0018] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0019] The present application provides a method and system for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture. The method first collects the initial state information of the autonomous driving vehicle and the obstacle data in the driving environment. Subsequently, a corresponding path planning model is constructed based on the acquired initial state of the vehicle and the obstacle information. With the dual support of the path planning model and the electronic and electrical architecture model, we use an optimized A* algorithm to accurately plan the global driving path of the autonomous driving vehicle. In order to further ensure the adaptability and real-time performance of the path, the present application also introduces a heuristic artificial potential field method, which can implement local dynamic path planning based on the latest dynamic information of the external environment of the autonomous driving vehicle. Finally, combining the results of global path planning and local dynamic planning, path tracking control is implemented on the autonomous driving vehicle, thereby achieving optimization, safety and smoothness of path planning while ensuring real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A flowchart of a method for path planning for an autonomous driving vehicle based on an electronic and electrical information architecture is provided for one embodiment of the present application.

[0022] Figure 2 A schematic diagram of an electronic and electrical architecture model provided for an autonomous driving vehicle according to an embodiment of the present application.

[0023] Figure 3 A framework diagram of a real-time motion planning method for an autonomous driving vehicle supported by an electronic and electrical architecture provided in one embodiment of the present application.

[0024] Figure 4A schematic diagram of the functional modules of an autonomous driving vehicle path planning system based on an electronic and electrical information architecture provided in accordance with one embodiment of the present application. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0027] Embodiment 1

[0028] like Figure 1 As shown, this embodiment provides a method for autonomous driving vehicle path planning based on electronic and electrical information architecture, including:

[0029] Step 101: Obtaining the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment;

[0030] Step 102: Establishing a path planning model according to the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; the path planning model is composed of factors including the number of path points of the autonomous driving vehicle, the area occupied by the path points, the drivable area, and the impassable area;

[0031] Step 103: Based on the path planning model and the electronic and electrical architecture model, an improved A* algorithm is used to perform global path planning for the autonomous driving vehicle; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network connection system, an autonomous driving domain controller, a switch, and a vehicle chassis domain controller; the perception system is used to perceive dynamic information of the external environment of the autonomous driving vehicle; the positioning system is used to receive the positioning information of the autonomous driving vehicle and the global scene map; the intelligent network connection system is used to receive information in the entire traffic network; the autonomous driving domain controller is used to execute the intelligent driving function of the autonomous driving vehicle; the switch is used to complete data interaction with different domain controllers; the vehicle chassis domain controller is used to execute chassis operation control of the autonomous driving vehicle;

[0032] Step 104: Based on the global path planning and the latest dynamic information of the external environment of the autonomous driving vehicle, a heuristic artificial potential field method is used to perform local dynamic planning of the autonomous driving vehicle;

[0033] Step 105: Perform path tracking control on the autonomous driving vehicle according to the global path planning and local dynamic planning of the autonomous driving vehicle.

[0034] In some embodiments, when executing step 101, the specific steps may be as follows:

[0035] Obtain the starting point position coordinates and the target point position coordinates of the autonomous driving vehicle and the number and area of ​​obstacles in the driving environment.

[0036] Specifically, the autonomous vehicle is regarded as a rigid body occupying a certain area. In the motion planning layer model, the state variable is S pln =(x sta ,y sta ,x tar ,y tar ,Obs1,Obs2,…Obs k-1 ,Obs k ), where (x sta ,y sta ) and (x tar ,y tar ) are the position coordinates of the starting point and the target point, k is the number of obstacles in the driving environment, Obs k ={(x obs j ,y obs j )|j=1,2,…n} k Represents the kth obstacle information composed of a large number of obstacle pixels.

[0037] When step 102 is executed, the specific steps may be as follows:

[0038] When performing path planning, it is necessary to model the path planning problem, as follows:

[0039] The planned path is denoted as P pln ={(x i ,f(x i ))|i=0,1,2,…m}, where m is the number of path points.

[0040] Where f(x) contains a series of specially constructed polynomials, each of which corresponds to a local segment of the entire path. The occupied area of ​​the i-th path point can be expressed as pln (i) indicates that it contains the actual occupied area and the safety boundary area. The safety boundary area is the outward extension of the actual occupied area, ensuring that the inaccessible area will not be crossed when searching for the next waypoint.

[0041] The drivable area of ​​the entire external environment of the autonomous vehicle is represented by Bsafe Indicates that the area cannot be passed by B unsafe =R 2 \B safe Indicates that R 2 That is the entire two-dimensional space.

[0042] The cost function of the state and the planned path is defined as J plan (S pln ,P pln ), then the model for solving the optimal solution of path planning can be seen as follows:

[0043]

[0044] Among them, A pln (i) is the occupied area of ​​the i-th path point, which is determined by the information of the i-th path point. pln It represents the radius of the safety margin area in the path planning process, which is determined by the single-step search distance d search and the maximum search step length Step max Decide.

[0045] In some embodiments, when executing step 103, the specific steps may be as follows:

[0046] Based on the traditional A* algorithm and search strategy, an improved variable step-size A* algorithm is constructed;

[0047] Construct a planning algorithm heuristic function based on the accumulated path cost of the autonomous driving vehicle at coordinate node n and the Euclidean distance from coordinate node n to the target point;

[0048] Based on the improved variable step-size A* algorithm and the planning algorithm heuristic function, the robust local weighted regression smoothing method is used to process discrete path points and obtain a smooth global path planning.

[0049] Specifically, the electronic and electrical architecture of self-driving cars belongs to a typical network-physical system, which realizes the connection and interaction between the network side and the physical side. In terms of the network system, the electronic and electrical architecture includes the perception system, positioning system, intelligent network system, switch, vehicle network and multiple domain controller units (DCU). In terms of the physical system, the architecture includes vehicle mechanical parts, such as chassis actuators, etc.

[0050] Among the components involved in this embodiment, the perception system is responsible for perceiving the dynamic information of the external environment; the positioning system is responsible for receiving the positioning information of the vehicle and the global scene map; the intelligent network connection system is responsible for receiving information in the entire traffic network (including other vehicles, traffic flow, etc.); the autonomous driving domain controller (ADSDCU) is responsible for realizing various intelligent driving functions of the autonomous driving vehicle, including the real-time path planning function; the switch, as the data exchange center in the electronic and electrical architecture, completes the interaction between different domain controllers; the vehicle chassis domain controller (Chassis DCU) is responsible for realizing the operation control of the chassis of the autonomous driving vehicle, and can convert the received control command into a signal that can be recognized by the actuator unit. The real-time path planning method proposed in this embodiment is integrated in the autonomous driving domain controller to achieve the global optimal and local dynamic real-time path planning goals. Data transmission between different domain controllers, between the perception unit and the domain controller, and between the domain controller and the mechanical execution system is realized through Ethernet or CAN lines.

[0051] The model of the electronic and electrical architecture mounted on the autonomous driving vehicle in this embodiment is as follows: Figure 2 As shown. The perception system, positioning system, and networking system transmit the information required for vehicle path planning (external environment information, scene map, vehicle coordinates, etc.) to the autonomous driving domain controller (ADSDCU) via Ethernet. The path planning method integrated in the ADSDCU performs global optimal and local dynamic path planning based on the received information, and then transmits the solved path to the switch via Ethernet, and uses the transmission capacity of the switch to send it to the vehicle chassis domain controller (Chassis DCU). Finally, the complete path planning process supported by the electronic and electrical architecture is completed.

[0052] Among them, Figure 3 As shown, based on the path planning model and the electronic and electrical architecture model, the improved smooth A* algorithm is used to implement global path planning, which can be specifically as follows:

[0053] 1) Research on variable step-size search A* algorithm:

[0054] When dealing with planning problems on large maps, the efficiency of the traditional A* algorithm is limited by the number of extension points. In the same search space, the number of extension points is directly related to the time complexity. In order to improve computational efficiency, this embodiment designs an improved variable step size A* algorithm. For example, when the current search position is in an open area, the search strategy can be more aggressive and the search step size can be set larger. On the contrary, when the current search position is surrounded by obstacles, the search strategy can be more conservative and the search step size can be set smaller. The search with a variable step size can be expressed as the following formula:

[0055]

[0056] Where d is the shortest distance from the current position to the obstacle. max and R min Represent the maximum search radius and the minimum search radius respectively. max and Step min Represents the maximum search step and the minimum search step respectively. The function round(x) represents taking the integer value of the input x.

[0057] 2) Construction of planning algorithm heuristic function:

[0058] The heuristic function of the planning algorithm in this embodiment is as follows:

[0059] f(n)=g(n)+h(n).

[0060] Among them, g(n) represents the cumulative path cost at coordinate node n, and h(n) represents the Euclidean distance from coordinate node n to the target point.

[0061] Based on the characteristics of graph search algorithms such as the A* algorithm, the path planning process is carried out based on the particle model of the autonomous driving car that contains location information. Since the actual occupied area and the safety boundary area have been introduced before, the safety and reliability of the path planning can be guaranteed.

[0062] 3) Robust local weighted regression smoothing:

[0063] Among them, the initial path obtained by the variable step size A* algorithm proposed in 1) and 2) is a series of discrete path points, and the smoothness of the path cannot be guaranteed. This embodiment uses a robust local weighted regression smoothing method to process discrete path points. The discrete points of the entire path are segmented and fitted. For each segmented path segment, the path points are sorted in time series, and the fitting function is as follows:

[0064]

[0065] Among them, α j is the coefficient of the jth term. Then, the optimization goal is to minimize the following formula:

[0066]

[0067] In the formula, (x i ,y i ) is the i-th initial point of the path segment.

[0068] K(x,x i ) represents the distance weight function, which is described as follows:

[0069]

[0070] Where D(t) is a cubic kernel function, and its expression is:

[0071]

[0072] The minimization formula mentioned above can be converted into a matrix expression:

[0073]

[0074] Then, the coefficients of the polynomial can be derived as:

[0075] A=(XTKX)-1XTKY.

[0076] At this stage, the initial polynomial f is formed fitting (x). The polynomial f obtained by the first local weighted regression fitting (x), the residual error of the fit can be calculated by the following formula:

[0077] ei=yi-ffitting(xi).

[0078] The residual weights are calculated using the bisquare function:

[0079] δi=B(ei / κ·s).

[0080] Where s is the median of the absolute value of the residual, and κ is the outlier factor. B(t) represents the bisquare kernel function, which is expressed as:

[0081]

[0082] The above function divides the waypoints into different levels according to the degree of outliers. Points that are far beyond the average degree of outliers are judged as sharp waypoints. Assigning the residual weight of sharp path points to 0 can avoid the formation of sharp corners. Then, the residual weight δ i Multiply by the distance weight function K(x,x i ), repeat the above local weighted regression process, and finally get the smooth global optimal path.

[0083] In some embodiments, when executing steps 104-105, the specific steps may be as follows:

[0084] like Figure 3 As shown in the figure, based on the planned global path and the external environment update information, the heuristic artificial potential field method is used to implement local dynamic planning, including:

[0085] 1) Construction of global heuristic potential field:

[0086] Based on the global path derived from the improved smooth A* algorithm, the global heuristic potential field is generated by the following formula:

[0087]

[0088] Among them, P smooth is the global optimal path after smoothing, (x PF ,y PF ) is the coordinate of any position in the global field. ξ represents the heuristic gain constant, p ref is the distance function from the position coordinate to the global path. ξ determines the gradient change rate of the global heuristic potential field. The larger the ξ value, the greater the change of the global heuristic potential field per unit distance.

[0089] 2) Construction of attractive potential field and repulsive potential field:

[0090] Then, based on the global heuristic potential field, the attractive potential field and the repulsive potential field are established. The expression of the attractive potential field is:

[0091]

[0092] Where τ is the attraction gain constant, p tar is the distance function from the current position coordinate to the target point. τ determines the gradient change rate of the attractive potential field. The larger the τ value, the greater the change of the attractive potential field per unit distance. The expression of the repulsive potential field is:

[0093]

[0094] Where η is the repulsion gain constant, p obs is the distance function from the current position coordinate to the obstacle, d obs is the influence range of the obstacle. η determines the gradient change rate of the repulsive potential field. The larger the η value, the greater the change of the repulsive potential field per unit distance.

[0095] 3) Construction of the total artificial potential field:

[0096] Then, the total artificial potential field is the vector superposition of the global heuristic potential field, the attractive potential field and the repulsive potential field. By reasonably setting the values ​​of ξ, τ and η, the expected interaction and superposition effect of the three potential fields can be achieved. The total artificial potential field expression is:

[0097]

[0098] Global heuristic potential field U global It is pre-generated based on the existing global scene map and is fixed. dynamic It is generated in real time based on changes in the local external environment, such as the sudden appearance of obstacles.

[0099] 4) Dynamic update planning of local paths:

[0100] Based on the total potential field calculated in 3), the real-time motion trajectory of the autonomous driving vehicle in the local environment is obtained by applying the gradient solution as follows:

[0101]

[0102] Assuming that the external environment is fixed, the autonomous vehicle can travel along the global optimal path planned by the improved smooth A* algorithm. Assuming that the optimal path is blocked due to sudden changes in the external environment, the autonomous vehicle can generate a local path planned in real time through the heuristic potential field method proposed in this embodiment to avoid sudden obstacles. Finally, the local dynamic path is output in real time from the path planning system and is continuously updated based on the perception information received in real time.

[0103] Then, according to the global path planning and local dynamic planning of the autonomous driving vehicle, the path tracking control is performed on the autonomous driving vehicle. Specifically, the information such as external environment changes, scene maps, vehicle coordinates, dynamic traffic flow, etc. are input into the path planning system in real time, and the global optimal path and local dynamic path are output to the vehicle chassis as the subsequent driving reference of the autonomous driving vehicle through the electronic and electrical architecture.

[0104] Embodiment 2

[0105] like Figure 4 As shown, this embodiment provides an autonomous driving vehicle path planning system based on an electronic and electrical information architecture, including:

[0106] The data acquisition module 401 is used to acquire the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment;

[0107] A path planning model building module 402 is used to establish a path planning model according to the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; the path planning model is composed of the number of path points of the autonomous driving vehicle, the area occupied by the path points, the drivable area and the impassable area;

[0108] The global path planning construction module 403 is used to perform global path planning for the autonomous driving vehicle based on the path planning model and the electronic and electrical architecture model using an improved A* algorithm; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network connection system, an autonomous driving domain controller, a switch, and a vehicle chassis domain controller; the perception system is used to perceive dynamic information of the external environment of the autonomous driving vehicle; the positioning system is used to receive the positioning information of the autonomous driving vehicle and the global scene map; the intelligent network connection system is used to receive information in the entire traffic network; the autonomous driving domain controller is used to execute the intelligent driving function of the autonomous driving vehicle; the switch is used to complete data interaction with different domain controllers; the vehicle chassis domain controller is used to execute chassis operation control of the autonomous driving vehicle;

[0109] A local dynamic planning module 404 is used to perform local dynamic planning of the autonomous driving vehicle using a heuristic artificial potential field method based on the global path planning and the latest dynamic information of the external environment of the autonomous driving vehicle;

[0110] The path tracking module 405 is used to perform path tracking control on the autonomous driving vehicle according to the global path planning and local dynamic planning of the autonomous driving vehicle.

[0111] In summary, this application has the following technical effects:

[0112] 1) For the real-time motion planning problem of autonomous vehicles supported by electronic and electrical architecture, the method proposed in this application has the significant advantage of overcoming the limitations of traditional graph search path planning algorithms (such as the A* algorithm) in dealing with real-time changes in scene maps, while solving the problem that artificial potential field methods are prone to falling into local optimal solutions in multi-obstacle environments. This method aims to achieve the goal of real-time path planning that emphasizes both global optimization and local updates, while ensuring real-time performance while taking into account the optimality, safety, and smoothness of the planning.

[0113] 2) This application builds an electronic and electrical architecture model designed specifically for autonomous vehicles, which realizes an integrated and centralized operation mode from real-time perception and positioning of the external environment, to the formulation of global and local path planning, and then to the execution of instructions by the vehicle chassis. This mode effectively ensures the real-time nature of the planning and decision-making process of autonomous vehicles.

[0114] 3) This application introduces an improved A* algorithm, which significantly improves the planning efficiency in large-scale scene maps by adopting a variable step size strategy. At the same time, combined with the robust local weighted regression smoothing technology, it effectively solves the problem of discrete path points, thereby generating a global path that is both optimal and smooth.

[0115] 4) This application proposes a heuristic artificial potential field method, which successfully avoids the shortcomings of the traditional artificial potential field method that is prone to fall into the local optimal solution by introducing the concept of global optimal heuristic reference, while retaining its original advantages such as high efficiency and strong real-time performance. This method can generate a path planning solution that meets the requirements of global optimality and can achieve local updates.

[0116] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0117] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for autonomous driving vehicle path planning based on electronic and electrical information architecture, characterized in that: include: Obtaining the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; A path planning model is established according to the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; the path planning model is composed of factors including the number of path points of the autonomous driving vehicle, the area occupied by the path points, the drivable area and the impassable area; Based on the path planning model and the electronic and electrical architecture model, an improved A* algorithm is used to perform global path planning for the autonomous driving vehicle; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network connection system, an autonomous driving domain controller, a switch, and a vehicle chassis domain controller; the perception system is used to perceive dynamic information of the external environment of the autonomous driving vehicle; the positioning system is used to receive positioning information of the autonomous driving vehicle and a global scene map; The intelligent network connection system is used to receive information from the entire traffic network; the autonomous driving domain controller is used to perform the intelligent driving function of the autonomous driving vehicle; the switch is used to complete data interaction with different domain controllers; the vehicle chassis domain controller is used to perform chassis operation control of the autonomous driving vehicle; Based on the global path planning and the latest dynamic information of the external environment of the autonomous driving vehicle, a heuristic artificial potential field method is used to perform local dynamic planning of the autonomous driving vehicle; The autonomous driving vehicle is controlled by path tracking according to the global path planning and local dynamic planning of the autonomous driving vehicle.

2. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 1, characterized in that: Obtain the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment, including: Obtain the starting point position coordinates and the target point position coordinates of the autonomous driving vehicle and the number and area of ​​obstacles in the driving environment.

3. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 1, characterized in that: The formula expression of the path planning model is: Among them, S pln =(x sta ,y sta ,x tar ,y tar ,Obs1,Obs2,...Obs k-1 ,Obs k ), P pln is the planned path, m is the number of path points, i is the i-th path point, i = 0, 1, 2, ... m, x sta and sta Represents the two-dimensional coordinates of the starting point, x tar and tar represents the two-dimensional coordinates of the target end point, Obs is the external environment obstacle information, k is the number of obstacles in the driving environment, f(x i ) is a polynomial containing several specific structures, each of which corresponds to a local segment of the entire path, B safe is the drivable area, B unsafe A is an impassable area. pln (i) is the occupied area of ​​the i-th path point, R pln Indicates the radius of the safety margin area during path planning, d search is the single-step search distance, Step max is the maximum search step length.

4. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 3, characterized in that: The solution formula for the impassable area is: B unsafe =R 2 \B safe ; Among them, R 2 for the entire two-dimensional space.

5. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 1, characterized in that: Based on the path planning model and the electronic and electrical architecture model, an improved A* algorithm is used to perform global path planning for the autonomous driving vehicle, specifically including: Based on the traditional A* algorithm and search strategy, an improved variable step-size A* algorithm is constructed; Construct a planning algorithm heuristic function based on the accumulated path cost of the autonomous driving vehicle at coordinate node n and the Euclidean distance from coordinate node n to the target point; Based on the improved variable step-size A* algorithm and the planning algorithm heuristic function, the robust local weighted regression smoothing method is used to process discrete path points and obtain a smooth global path planning.

6. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 5, characterized in that: The improved variable step-size A* algorithm is specifically: Where d is the shortest distance from the current position to the obstacle; R max and R min Respectively represent the maximum search radius and the minimum search radius; Step max and Step min They represent the maximum search step size and the minimum search step size respectively; the function round(x) represents taking the integer value of the input x.

7. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 5, characterized in that: The planning algorithm heuristic function is specifically: f(n)=g(n)+h(n); Among them, g(n) represents the cumulative path cost at coordinate node n, and h(n) represents the Euclidean distance from coordinate node n to the target point.

8. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 5, characterized in that: Based on the improved variable step size A* algorithm and the planning algorithm heuristic function, the robust local weighted regression smoothing method is used to process discrete path points to obtain a smooth global path planning, which includes: According to the formula Perform piecewise fitting on discrete path points; A robust local weighted regression smoothing method is used for each segmented path segment to determine the minimum objective function Among them, (x i ,y i ) is the i-th initial point of the path segment, K(x,x i ) represents the distance weight function, K(x,x i )=D(t)*B(t), D(t) is the cubic kernel function, B(t) is the bisquare kernel function; The initial polynomial f is obtained by minimizing the objective function fitting (x); Calculate the initial polynomial f fitting The residual weight δ is calculated using the bisquare function according to the median s of the absolute value of the residual and the outlier factor κ. i ; The residual weight δ i Multiply by the distance weight function K(x,x i ), and repeat the local weighted regression process to obtain a smooth global optimal path.

9. The method for path planning of an autonomous driving vehicle based on an electronic and electrical information architecture according to claim 1, characterized in that: The formula of the heuristic artificial potential field method is: Among them, U global is the global heuristic potential field, (x PF ,y PF ) is the coordinate of any position in the global field, U dynamic is the dynamic potential field, Uatt is the attractive potential field, U rep Repulsive potential field.

10. A path planning system for an autonomous driving vehicle based on an electronic and electrical information architecture, characterized in that: include: A data acquisition module, used to obtain the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; A path planning model building module, used to establish a path planning model according to the initial vehicle state of the autonomous driving vehicle and obstacle information in the driving environment; the path planning model is composed of the number of path points of the autonomous driving vehicle, the area occupied by the path points, the drivable area and the impassable area; A global path planning construction module is used to perform global path planning for the autonomous driving vehicle based on the path planning model and the electronic and electrical architecture model using an improved A* algorithm; the electronic and electrical architecture model is composed of a perception system, a positioning system, an intelligent network connection system, an autonomous driving domain controller, a switch, and a vehicle chassis domain controller; the perception system is used to perceive dynamic information of the external environment of the autonomous driving vehicle; the positioning system is used to receive positioning information of the autonomous driving vehicle and a global scene map; The intelligent network connection system is used to receive information from the entire traffic network; the autonomous driving domain controller is used to perform the intelligent driving function of the autonomous driving vehicle; the switch is used to complete data interaction with different domain controllers; the vehicle chassis domain controller is used to perform chassis operation control of the autonomous driving vehicle; A local dynamic planning module, for performing local dynamic planning of the autonomous driving vehicle using a heuristic artificial potential field method based on global path planning and the latest dynamic information of the external environment of the autonomous driving vehicle; A path tracking module is used to perform path tracking control on the autonomous driving vehicle based on the global path planning and local dynamic planning of the autonomous driving vehicle.

Images