A vehicle automatic driving trajectory planning method and system based on edge computing
By adopting the trajectory planning method based on edge computing in the vehicle autonomous driving system, using artificial potential field method and genetic algorithm to optimize the vehicle trajectory, the safety risk problems caused by cloud computing delay are solved, and more efficient and safe autonomous driving trajectory planning is achieved. .
Patent Information
- Application Number
- CN202411060815.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-05
AI Technical Summary
In the prior art, the vehicle's autonomous driving trajectory planning relies on cloud computing, resulting in delays in data upload and calculation, affecting the timeliness of the vehicle's information reception, and increasing safety risks.
The vehicle's autonomous driving trajectory planning method is adopted based on edge computing. By receiving vehicle driving environment data, a driving environment potential field model is generated, and artificial potential field method is used to improve it. Combining model prediction control and genetic algorithm, vehicle trajectory planning is optimized.
Through edge computing, data transmission delay is reduced, the security and real-time nature of vehicle autonomous driving trajectory planning is improved, and the dependence on cloud computing centers is reduced.
Smart Images

Figure CN118579109B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle autonomous driving trajectory planning, and specifically to a vehicle autonomous driving trajectory planning method and system based on edge computing. Background Art
[0002] The autonomous driving system senses the road environment through the on-board sensor system, and controls the vehicle's steering and speed based on the road, vehicle position and obstacle information obtained, so that the vehicle can travel safely and reliably on the road and reach the predetermined location. The autonomous driving trajectory planning is to plan the optimal driving trajectory of the vehicle from the current position to the target position under the given road environment and vehicle dynamic constraints.
[0003] In the prior art, when planning the trajectory of an autonomous vehicle, a cloud storage computing method is usually used. The vehicle uploads the driving data to a cloud computing center. After the cloud computing center performs computing processing, the computing results are returned to the vehicle. However, in actual use, the data processing method of the cloud computing center has a high delay. It is necessary to upload the data and wait for the calculation before receiving the data. Since the driving environment on the road is relatively complex, when the cloud computing center has a computing delay, the vehicle will not receive the information in time, which may lead to a traffic accident. Summary of the invention
[0004] In order to address the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a vehicle autonomous driving trajectory planning method and system based on edge computing, which can plan the vehicle autonomous driving trajectory and improve the safety of vehicle autonomous driving trajectory planning.
[0005] In the first aspect, the purpose of the present invention can be achieved by the following technical solution: a vehicle automatic driving trajectory planning method based on edge computing, the method comprising the following steps:
[0006] Receiving vehicle driving environment related data, generating a vehicle driving environment potential field model according to the vehicle driving environment related data, and improving the vehicle driving environment potential field model by using an artificial potential field method to obtain an improved traditional vehicle driving environment model;
[0007] The vehicle is set as a mass point, the driving state of the vehicle is obtained based on the mass point, the driving trajectory of the vehicle is obtained, the driving trajectory of the vehicle is input into the improved traditional vehicle driving environment model, and the optimal driving trajectory after optimization is output, wherein the improved traditional vehicle driving environment model is solved based on a pre-established model predictive control and artificial potential field method vehicle trajectory planner, and the pre-established model predictive control and artificial potential field method vehicle trajectory planner performs parameter optimization based on a genetic algorithm.
[0008] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the vehicle driving environment potential field model includes: a road boundary potential field model, a lane line potential field model and an environment vehicle potential field model.
[0009] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: the vehicle driving environment potential field model includes: a road boundary potential field model, a lane line potential field model and an environmental vehicle potential field model, and the road boundary potential field value in the road boundary potential field model is greater than the lane line potential field value of the lane line potential field model.
[0010] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the road boundary potential field model construction formula is:
[0011]
[0012] is the road boundary potential field function, where is the road boundary potential field coefficient, is the horizontal coordinate of the road boundary line, represents the i-th road boundary, is the current position of the vehicle, is the road boundary potential field coefficient;
[0013] The formula for constructing the lane line potential field model is:
[0014]
[0015] is the lane line potential field function, where is the lane line potential field coefficient, is the horizontal line position of the i-th lane line, is the lane line convergence coefficient, It indicates the boundary of the i-th road;
[0016] When modeling the environmental vehicle potential field model: to ensure the driving safety of the vehicle and maintain a safe driving distance as a constraint, the lateral and longitudinal potential field models of the environmental vehicle are constructed, wherein the safety distance of the environmental vehicle potential field model depends on the driving speed of the environmental vehicle potential field model. The faster the driving speed, the larger the safety distance is set; the slower the driving speed, the smaller the safety distance is set;
[0017] The vehicle driving environment potential field model is constructed by combining the road boundary potential field model, the lane line potential field model and the environmental vehicle potential field modeling.
[0018] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: when the artificial potential field method is used to improve the vehicle driving environment potential field model, road boundaries, lane lines and environmental vehicle potential field models are respectively established. The road boundaries, lane lines and environmental vehicle potential field models are used for the vehicle to travel along the center line when driving and not changing lanes, and it is prohibited to drive out of the boundary line.
[0019] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the improved traditional vehicle driving environment model is solved based on a vehicle trajectory planner of a pre-established model predictive control and artificial potential field method, including:
[0020] Generate safe operation trajectory according to vehicle motion status and design overtaking trajectory optimization strategy;
[0021] Genetic algorithm is used to optimize the parameters of the vehicle trajectory planner based on the pre-established model predictive control and artificial potential field method, and the optimized optimal driving trajectory is output.
[0022] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: the optimal driving trajectory includes: road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness assessment and lateral acceleration increment.
[0023] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the road centerline tracking objective function is:
[0024]
[0025] The objective function of the environmental vehicle potential field is:
[0026]
[0027] The objective function of the potential field in the feasible overtaking area is:
[0028]
[0029] The objective function of the road boundary potential field is:
[0030]
[0031] The objective function of trajectory steepness evaluation is:
[0032]
[0033] The lateral acceleration increment objective function is:
[0034]
[0035] in 、 、 、 、 、 They are the target item weights of road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation, and lateral acceleration increment. 、 、 、 are target values, represents the i-th road boundary, , For time, is the acceleration, is the trajectory steepness penalty, i.e., the difference between the vehicle position and the vehicle offset angle;
[0036] The function formula to be solved is:
[0037]
[0038] represents the objective function, represents a linear inequality constraint, represents a linear equality constraint, and Represents upper and lower boundary constraints, represents a nonlinear constraint function.
[0039] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: the pre-established vehicle trajectory planner of model predictive control and artificial potential field method performs parameter optimization based on a genetic algorithm to obtain the optimal driving trajectory, including: solving the improved traditional vehicle driving environment model to obtain the optimal lateral acceleration, using curve fitting to process the optimal lateral acceleration discrete sequence, and using a cross calculation method to optimize the model predictive control and artificial potential field method, so that the model predictive control rolling optimization mechanism in the pre-established vehicle trajectory planner of model predictive control and artificial potential field method is combined with the potential field model of the artificial potential field method, and finally solving the vehicle trajectory planner to obtain the optimal solution, and the vehicle trajectory planner with the optimal solution solves the improved traditional vehicle driving environment model to obtain the optimal driving trajectory.
[0040] In a second aspect, in order to achieve the above-mentioned purpose, the present invention discloses a vehicle automatic driving trajectory planning system based on edge computing, comprising:
[0041] An improved model module is used to receive vehicle driving environment related data, generate a vehicle driving environment potential field model according to the vehicle driving environment related data, and improve the vehicle driving environment potential field model using an artificial potential field method to obtain an improved traditional vehicle driving environment model;
[0042] The trajectory optimization module is used to set the vehicle as a mass point, obtain the vehicle's driving state based on the mass point, obtain the vehicle's driving trajectory, input the vehicle's driving trajectory into the improved traditional vehicle driving environment model, and output the optimized optimal driving trajectory, wherein the improved traditional vehicle driving environment model is solved based on a pre-established model predictive control and artificial potential field method vehicle trajectory planner, and the pre-established model predictive control and artificial potential field method vehicle trajectory planner performs parameter optimization based on a genetic algorithm.
[0043] In combination with the second aspect, in some implementations of the second aspect, the system further includes: the vehicle driving environment potential field model in the improved model module includes: a road boundary potential field model, a lane line potential field model and an environment vehicle potential field model, and the road boundary potential field value in the road boundary potential field model is greater than the lane line potential field value of the lane line potential field model;
[0044] The formula for constructing the road boundary potential field model in the improved model module is:
[0045]
[0046] is the road boundary potential field function, where is the road boundary potential field coefficient, is the horizontal coordinate of the road boundary line, represents the i-th road boundary, is the current position of the vehicle, is the road boundary potential field coefficient;
[0047] The formula for constructing the lane line potential field model is:
[0048]
[0049] is the lane line potential field function, where is the lane line potential field coefficient, is the horizontal line position of the i-th lane line, is the lane line convergence coefficient, It indicates the boundary of the i-th road;
[0050] When modeling the environmental vehicle potential field model: to ensure the driving safety of the vehicle and maintain a safe driving distance as a constraint, the lateral and longitudinal potential field models of the environmental vehicle are constructed, wherein the safety distance of the environmental vehicle potential field model depends on the driving speed of the environmental vehicle potential field model. The faster the driving speed, the larger the safety distance is set; the slower the driving speed, the smaller the safety distance is set;
[0051] The vehicle driving environment potential field model is constructed by combining the road boundary potential field model, the lane line potential field model and the environmental vehicle potential field modeling;
[0052] When the artificial potential field method is used to improve the vehicle driving environment potential field model in the improved model module, the road boundary, lane line and environmental vehicle potential field models are respectively established. The road boundary, lane line and environmental vehicle potential field models are used for the vehicle to drive along the center line when driving and not changing lanes, and it is forbidden to drive out of the boundary line;
[0053] The trajectory optimization module improves the traditional vehicle driving environment model and solves the vehicle trajectory planner based on the pre-established model predictive control and artificial potential field method, including:
[0054] Generate safe operation trajectory according to vehicle motion status and design overtaking trajectory optimization strategy;
[0055] Genetic algorithm is used to optimize the parameters of the vehicle trajectory planner of the pre-established model predictive control and artificial potential field method, and the optimized optimal driving trajectory is output;
[0056] The optimal driving trajectory in the trajectory optimization module includes: road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation and lateral acceleration increment;
[0057] The objective function of road centerline tracking in the trajectory optimization module is:
[0058]
[0059] The objective function of the environmental vehicle potential field is:
[0060]
[0061] The objective function of the potential field in the feasible overtaking area is:
[0062]
[0063] The objective function of the road boundary potential field is:
[0064]
[0065] The objective function of trajectory steepness evaluation is:
[0066]
[0067] The lateral acceleration increment objective function is:
[0068]
[0069] in 、 、 、 、 、 They are the target item weights of road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation, and lateral acceleration increment. 、 、 、 are target values, represents the i-th road boundary, , For time, is the acceleration, is the trajectory steepness penalty, i.e., the difference between the vehicle position and the vehicle offset angle;
[0070] The function formula to be solved is:
[0071]
[0072] represents the objective function, represents a linear inequality constraint, represents a linear equality constraint, and Represents upper and lower boundary constraints, represents a nonlinear constraint function;
[0073] The vehicle trajectory planner of the model predictive control and artificial potential field method pre-established in the trajectory optimization module performs parameter optimization based on the genetic algorithm to obtain the process of the optimal driving trajectory, including: obtaining the optimal lateral acceleration by solving the improved traditional vehicle driving environment model, processing the optimal lateral acceleration discrete sequence by curve fitting, optimizing the model predictive control and the artificial potential field method by cross calculation, combining the model predictive control rolling optimization mechanism in the pre-established vehicle trajectory planner of the model predictive control and artificial potential field method with the potential field model of the artificial potential field method, and finally solving the vehicle trajectory planner to obtain the optimal solution, and the vehicle trajectory planner with the optimal solution solves the improved traditional vehicle driving environment model to obtain the optimal driving trajectory.
[0074] Beneficial effects of the present invention:
[0075] The present invention uses a vehicle planner and an edge computing trajectory planning platform to conduct simulation experiments. After testing and verification, it is found that the vehicle trajectory of edge computing is feasible. The interaction between the edge device and the edge computing server can avoid the defects of computing delays in the cloud computing center, thereby improving the security of vehicle automatic driving trajectory planning. By adopting the edge computing vehicle automatic driving planning method, the traditional vehicle driving model is optimized, improved by the artificial potential field method, and the corresponding potential field model is built. The function is solved, and the point mass model is used to depict the driving state of the vehicle. The vehicle trajectory is optimized based on model predictive control and the artificial potential field method, thereby realizing the function of planning the vehicle automatic driving trajectory. BRIEF DESCRIPTION OF THE DRAWINGS
[0076] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art description are briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0077] Figure 1 It is a schematic flow chart of the method of the present invention;
[0078] Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION
[0079] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0080] Embodiment 1:
[0081] The following is an introduction to the relevant terms involved in the embodiments of the present application:
[0082] Edge computing is a computing model that focuses on moving computing power and storage resources closer to users or data sources, thereby reducing data transmission delays and improving processing efficiency.
[0083] In edge computing systems, computing and storage capabilities are combined on devices, sensors, and edge nodes at the edge of the network, which can include IoT sensors, smartphones, computers, etc. This distributed computing model can provide faster, more efficient, and more secure computing services, which is particularly important in application scenarios that process real-time data and require low latency.
[0084] Compared with cloud computing, which is concentrated in remote data centers, edge computing transfers part of the computing tasks to the edge of the network, so that data processing and analysis can be performed closer to the location where the data is generated. This not only reduces the delay in data transmission, but also reduces dependence on centralized cloud servers.
[0085] Genetic Algorithm: Genetic Algorithm (GA) is designed and proposed based on the laws of evolution of organisms in nature. It is a computational model of biological evolution that simulates the natural selection and genetic mechanism of biological evolution. It is a method of searching for the optimal solution by simulating the natural evolution process. The algorithm uses mathematical methods and computer simulation operations to convert the problem-solving process into a process similar to the crossover and mutation of chromosome genes in biological evolution. When solving more complex combinatorial optimization problems, it can usually obtain better optimization results faster than some conventional optimization algorithms. Genetic algorithms have been widely used in combinatorial optimization, machine learning, signal processing, adaptive control, artificial life and other fields.
[0086] like Figure 1 As shown, a vehicle automatic driving trajectory planning method based on edge computing is characterized in that the method includes the following steps:
[0087] S101: receiving vehicle driving environment related data, generating a vehicle driving environment potential field model according to the vehicle driving environment related data, and improving the vehicle driving environment potential field model using an artificial potential field method to obtain an improved traditional vehicle driving environment model;
[0088] The vehicle driving environment potential field model includes: a road boundary potential field model, a lane line potential field model and an environment vehicle potential field model, and the road boundary potential field value in the road boundary potential field model is greater than the lane line potential field value in the lane line potential field model;
[0089] The formula for constructing the road boundary potential field model is:
[0090]
[0091] is the road boundary potential field function, where is the road boundary potential field coefficient, is the horizontal coordinate of the road boundary line, represents the i-th road boundary, is the current position of the vehicle, is the road boundary potential field coefficient.
[0092] The formula for constructing the lane line potential field model is:
[0093]
[0094] is the lane line potential field function, where is the lane line potential field coefficient, is the horizontal line position of the i-th lane line, is the lane line convergence coefficient, It represents the boundary of the i-th road.
[0095] When modeling the environmental vehicle potential field model: ensure the driving safety of the vehicle, maintain a safe driving distance, and build the lateral and longitudinal potential field models of the environmental vehicle, wherein the safe distance of the environmental vehicle potential field model depends on the driving speed of the environmental vehicle potential field model.
[0096] When using the artificial potential field method for improvement, it is necessary to set up the road boundary, lane line and environmental vehicle potential field model respectively. When the vehicle is driving, it drives along the center line without changing lanes and is prohibited from driving out of the boundary line.
[0097] S102: setting the vehicle as a mass point, obtaining the driving state of the vehicle based on the mass point, obtaining the driving trajectory of the vehicle, inputting the driving trajectory of the vehicle into an improved traditional vehicle driving environment model, and outputting the optimized optimal driving trajectory, wherein the improved traditional vehicle driving environment model is solved based on a pre-established vehicle trajectory planner of model predictive control and artificial potential field method, and the pre-established vehicle trajectory planner of model predictive control and artificial potential field method performs parameter optimization based on a genetic algorithm.
[0098] Among them, when the vehicle trajectory planner based on the pre-established model predictive control and artificial potential field method is used to solve the improved traditional vehicle driving environment model, the improved traditional vehicle driving environment model is converted into the penalty term of the model predictive control objective function by utilizing the optimization mechanism of the model predictive control, and the vehicle trajectory planning problem is converted into an objective function solution.
[0099] Improve the traditional vehicle driving environment model based on the pre-established model predictive control and artificial potential field method vehicle trajectory planner to solve, including:
[0100] Generate safe operation trajectory according to vehicle motion status and design overtaking trajectory optimization strategy;
[0101] Specifically, the scheme of the present invention is further described below through an embodiment, and the overtaking trajectory optimization strategy is as follows:
[0102] The overtaking process is divided into three stages: two lane changes and one overtaking. According to the initial speed of the vehicle, the speed of the front vehicle and the distance from the front vehicle, the target speed, target position and lane change overtaking time of the vehicle are calculated to establish the objective function. Then, the hybrid particle swarm algorithm is used to solve the objective function, and the optimal lane change longitudinal displacement and lane change time under different vehicle speeds are obtained, and then the optimal overtaking trajectory of each stage is calculated. Finally, the vehicle overtaking trajectory tracking control is realized based on the model predictive control (MPC) method. According to the three-segment overtaking trajectory planned by the quintic polynomial curve, combined with the vehicle overtaking working conditions, the vehicle's collision avoidance constraints, lateral displacement constraints, control quantity constraints and other constraints are comprehensively considered to determine a reasonable objective function and design a trajectory tracking controller based on model predictive control. According to the actual overtaking experience, drivers will choose to accelerate to overtake the front vehicle when overtaking. Therefore, the reference speed is introduced when the vehicle overtakes, and the speed tracking controller based on fuzzy PI control is designed, which shows that the vehicle can track the target trajectory smoothly and accurately, and can complete the overtaking according to the reference speed.
[0103] Genetic algorithm is used to optimize the parameters of the vehicle trajectory planner based on the pre-established model predictive control and artificial potential field method, and the optimized optimal driving trajectory is output.
[0104] Specifically, the present invention is further described below by way of embodiments:
[0105] Parameter optimization process:
[0106] 1. Each individual represents each trajectory path;
[0107] 2. Using the shortest path and path safety as evaluation criteria, evaluate the fitness of each individual and select individuals with high fitness;
[0108] 3. Then carry out genetic operations, including crossover, mutation and other processes, to form new individuals;
[0109] 4. Then, the new population individuals are evaluated according to their fitness, and individuals with high fitness are selected to enter the second round of genetic operations;
[0110] 5. Repeat steps 2 to 4 until the predetermined fitness threshold is reached and the optimal solution is output.
[0111] Furthermore, the objective function includes: road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation and lateral acceleration increment.
[0112] The road centerline tracking objective function is:
[0113]
[0114] The objective function of the environmental vehicle potential field is:
[0115]
[0116] The objective function of the potential field in the feasible overtaking area is:
[0117]
[0118] The objective function of the road boundary potential field is:
[0119]
[0120] The objective function of trajectory steepness evaluation is:
[0121]
[0122] The lateral acceleration increment objective function is:
[0123]
[0124] in 、 、 、 、 、 They are the target item weights of road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation, and lateral acceleration increment. 、 、 、 is the target value of the item, represents the i-th road boundary, , For time, is the acceleration, is the trajectory steepness penalty, i.e., the difference between the vehicle position and the vehicle offset angle;
[0125] The function formula to be solved is:
[0126]
[0127] represents the objective function, represents a linear inequality constraint, represents a linear equality constraint, and Represents upper and lower boundary constraints, represents a nonlinear constraint function.
[0128] The process of optimizing parameters of the improved traditional vehicle driving environment model based on genetic algorithm to generate the optimal trajectory is as follows: the optimal vehicle acceleration is obtained by solving the problem, the discrete sequence is processed by curve fitting, and the model predictive control and the artificial potential field method are optimized by cross calculation. The rolling optimization mechanism of the model predictive control is combined with the potential field model of the artificial potential field method to obtain the optimal solution as the optimal trajectory.
[0129] Specifically, the present invention is further described below by way of embodiments:
[0130] Based on the edge computing architecture, an edge computing trajectory planning platform was built:
[0131] The overall design is carried out according to the performance characteristics of the platform, mainly including the vehicle-mounted edge device and edge computing server.
[0132] The on-board edge device covers human-computer interaction interface, vehicle positioning and map display, communication interface, data storage and information transmission modules.
[0133] The edge computing server includes human-computer interaction interface, communication interface, data analysis, data storage and map display modules.
[0134] Furthermore, simulation experiments and functional tests were carried out on the designed vehicle trajectory planner and edge computing trajectory planning platform, and three longitudinal simulation tests were designed to verify the optimization performance and effectiveness of vehicle trajectory planning.
[0135] Carry out lateral comparison tests at different relative speeds to verify the applicability of the vehicle trajectory planner.
[0136] Second, as Figure 2 As shown, in order to achieve the above-mentioned purpose, the present invention discloses a vehicle automatic driving trajectory planning system based on edge computing, comprising:
[0137] The improved model module 11 is used to receive vehicle driving environment related data, generate a vehicle driving environment potential field model according to the vehicle driving environment related data, and improve the vehicle driving environment potential field model using an artificial potential field method to obtain an improved traditional vehicle driving environment model;
[0138] The trajectory optimization module 12 is used to set the vehicle as a mass point, obtain the driving state of the vehicle based on the mass point, obtain the driving trajectory of the vehicle, input the driving trajectory of the vehicle into the improved traditional vehicle driving environment model, and output the optimized optimal driving trajectory, wherein the improved traditional vehicle driving environment model is solved based on a pre-established model predictive control and artificial potential field method vehicle trajectory planner, and the pre-established model predictive control and artificial potential field method vehicle trajectory planner performs parameter optimization based on a genetic algorithm.
[0139] In combination with the second aspect, in some implementations of the second aspect, the system further includes: the vehicle driving environment potential field model in the improved model module 11 includes: a road boundary potential field model, a lane line potential field model and an environment vehicle potential field model, and the road boundary potential field value in the road boundary potential field model is greater than the lane line potential field value of the lane line potential field model;
[0140] The formula for constructing the road boundary potential field model in the improved model module 11 is:
[0141]
[0142] is the road boundary potential field function, where is the road boundary potential field coefficient, is the horizontal coordinate of the road boundary line, represents the i-th road boundary, is the current position of the vehicle, is the road boundary potential field coefficient;
[0143] The formula for constructing the lane line potential field model is:
[0144]
[0145] is the lane line potential field function, where is the lane line potential field coefficient, is the horizontal line position of the i-th lane line, is the lane line convergence coefficient, It indicates the boundary of the i-th road;
[0146] When modeling the environmental vehicle potential field model: to ensure the driving safety of the vehicle and maintain a safe driving distance as a constraint, the lateral and longitudinal potential field models of the environmental vehicle are constructed, wherein the safety distance of the environmental vehicle potential field model depends on the driving speed of the environmental vehicle potential field model. The faster the driving speed, the larger the safety distance is set; the slower the driving speed, the smaller the safety distance is set;
[0147] The vehicle driving environment potential field model is constructed by combining the road boundary potential field model, the lane line potential field model and the environmental vehicle potential field modeling;
[0148] When the vehicle driving environment potential field model is improved by using the artificial potential field method in the improved model module 11, the road boundary, lane line and environment vehicle potential field models are respectively established. The road boundary, lane line and environment vehicle potential field models are used for the vehicle to drive along the center line when driving and not changing lanes, and it is forbidden to drive out of the boundary line;
[0149] The trajectory optimization module 12 improves the traditional vehicle driving environment model and solves the vehicle trajectory planner based on the pre-established model predictive control and artificial potential field method, including:
[0150] Generate safe operation trajectory according to vehicle motion status and design overtaking trajectory optimization strategy;
[0151] Genetic algorithm is used to optimize the parameters of the vehicle trajectory planner of the pre-established model predictive control and artificial potential field method, and the optimized optimal driving trajectory is output;
[0152] The optimal driving trajectory in the trajectory optimization module 12 includes: road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation and lateral acceleration increment;
[0153] The road centerline tracking objective function in the trajectory optimization module 12 is:
[0154]
[0155] The objective function of the environmental vehicle potential field is:
[0156]
[0157] The objective function of the potential field in the feasible overtaking area is:
[0158]
[0159] The objective function of the road boundary potential field is:
[0160]
[0161] The objective function of trajectory steepness evaluation is:
[0162]
[0163] The lateral acceleration increment objective function is:
[0164]
[0165] in 、 、 、 、 、 They are the target item weights of road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation, and lateral acceleration increment. 、 、 、 are target values, represents the i-th road boundary, , For time, is the acceleration, is the trajectory steepness penalty, i.e., the difference between the vehicle position and the vehicle offset angle;
[0166] The function formula to be solved is:
[0167]
[0168] represents the objective function, represents a linear inequality constraint, represents a linear equality constraint, and Represents upper and lower boundary constraints, represents a nonlinear constraint function;
[0169] The vehicle trajectory planner of the model predictive control and artificial potential field method pre-established in the trajectory optimization module 12 performs parameter optimization based on a genetic algorithm to obtain a process of the optimal driving trajectory, including: obtaining the optimal lateral acceleration by solving the improved traditional vehicle driving environment model, processing the optimal lateral acceleration discrete sequence by curve fitting, optimizing the model predictive control and the artificial potential field method by cross calculation, combining the model predictive control rolling optimization mechanism in the pre-established vehicle trajectory planner of the model predictive control and artificial potential field method with the potential field model of the artificial potential field method, and finally solving the vehicle trajectory planner to obtain the optimal solution, and the vehicle trajectory planner with the optimal solution solves the improved traditional vehicle driving environment model to obtain the optimal driving trajectory.
[0170] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.
[0171] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium, on which a computer program is stored, and the computer program is executed by the processor to execute the above method. The storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.
[0172] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0173] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure may have various changes and improvements, and these changes and improvements fall within the scope of the present disclosure to be protected.
Claims
1. A vehicle automatic driving trajectory planning method based on edge computing, characterized in that: The method comprises the following steps: Receiving vehicle driving environment related data, generating a vehicle driving environment potential field model according to the vehicle driving environment related data, and improving the vehicle driving environment potential field model by using an artificial potential field method to obtain an improved traditional vehicle driving environment model; The vehicle driving environment potential field model includes: a road boundary potential field model, a lane line potential field model and an environment vehicle potential field model, and the road boundary potential field value in the road boundary potential field model is greater than the lane line potential field value of the lane line potential field model; The construction formula of the road boundary potential field model is: is the road boundary potential field function, where is the road boundary potential field coefficient, is the horizontal coordinate of the road boundary line, represents the i-th road boundary, is the current position of the vehicle, is the road boundary potential field coefficient; The formula for constructing the lane line potential field model is: is the lane line potential field function, where is the lane line potential field coefficient, is the horizontal line position of the i-th lane line, is the lane line convergence coefficient, It indicates the boundary of the i-th road; When modeling the environmental vehicle potential field model: to ensure the driving safety of the vehicle and maintain a safe driving distance as a constraint, the lateral and longitudinal potential field models of the environmental vehicle are constructed, wherein the safety distance of the environmental vehicle potential field model depends on the driving speed of the environmental vehicle potential field model. The faster the driving speed, the larger the safety distance is set; the slower the driving speed, the smaller the safety distance is set; The vehicle driving environment potential field model is constructed by combining the road boundary potential field model, the lane line potential field model and the environmental vehicle potential field modeling; The vehicle is set as a mass point, the driving state of the vehicle is obtained based on the mass point, the driving trajectory of the vehicle is obtained, the driving trajectory of the vehicle is input into the improved traditional vehicle driving environment model, and the optimal driving trajectory after optimization is output, wherein the improved traditional vehicle driving environment model is solved based on a pre-established model predictive control and artificial potential field method vehicle trajectory planner, and the pre-established model predictive control and artificial potential field method vehicle trajectory planner is optimized based on a genetic algorithm; The improved traditional vehicle driving environment model is solved based on a vehicle trajectory planner of a pre-established model predictive control and an artificial potential field method, including: Generate safe operation trajectory according to vehicle motion status and design overtaking trajectory optimization strategy; Genetic algorithm is used to optimize the parameters of the vehicle trajectory planner of the pre-established model predictive control and artificial potential field method, and the optimized optimal driving trajectory is output; The optimal driving trajectory includes: road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation and lateral acceleration increment; The road centerline tracking objective function is: The objective function of the environmental vehicle potential field is: The objective function of the potential field in the feasible overtaking area is: The objective function of the road boundary potential field is: The objective function of trajectory steepness evaluation is: The lateral acceleration increment objective function is: in 、 、 、 、 、 They are the target item weights of road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation, and lateral acceleration increment. 、 、 、 are target values, represents the i-th road boundary, , For time, is the acceleration, is the trajectory steepness penalty, i.e., the difference between the vehicle position and the vehicle offset angle; The function formula to be solved is: represents the objective function, represents a linear inequality constraint, represents a linear equality constraint, and Represents upper and lower boundary constraints, represents a nonlinear constraint function.
2. The method for automatic driving trajectory planning of a vehicle based on edge computing according to claim 1, characterized in that: When the vehicle driving environment potential field model is improved by using the artificial potential field method, road boundaries, lane lines and environmental vehicle potential field models are respectively established. The road boundaries, lane lines and environmental vehicle potential field models are used for the vehicle to drive along the center line when driving and not changing lanes, and it is prohibited to drive out of the boundary line.
3. The method for automatic driving trajectory planning of a vehicle based on edge computing according to claim 1, characterized in that: The process of obtaining the optimal driving trajectory by optimizing parameters of the vehicle trajectory planner of the pre-established model predictive control and artificial potential field method based on a genetic algorithm includes: obtaining the optimal lateral acceleration by solving the improved traditional vehicle driving environment model, processing the optimal lateral acceleration discrete sequence by curve fitting, optimizing the model predictive control and the artificial potential field method by cross calculation, combining the model predictive control rolling optimization mechanism in the pre-established vehicle trajectory planner of the model predictive control and artificial potential field method with the potential field model of the artificial potential field method, finally solving the vehicle trajectory planner to obtain the optimal solution, and the vehicle trajectory planner with the optimal solution solves the improved traditional vehicle driving environment model to obtain the optimal driving trajectory.
4. A vehicle autonomous driving trajectory planning system based on edge computing, characterized in that: include: An improved model module is used to receive vehicle driving environment related data, generate a vehicle driving environment potential field model according to the vehicle driving environment related data, and improve the vehicle driving environment potential field model using an artificial potential field method to obtain an improved traditional vehicle driving environment model; The vehicle driving environment potential field model in the improved model module includes: a road boundary potential field model, a lane line potential field model and an environment vehicle potential field model, and the road boundary potential field value in the road boundary potential field model is greater than the lane line potential field value of the lane line potential field model; The formula for constructing the road boundary potential field model in the improved model module is: is the road boundary potential field function, where is the road boundary potential field coefficient, is the horizontal coordinate of the road boundary line, represents the i-th road boundary, is the current position of the vehicle, is the road boundary potential field coefficient; The formula for constructing the lane line potential field model is: is the lane line potential field function, where is the lane line potential field coefficient, is the horizontal line position of the i-th lane line, is the lane line convergence coefficient, It indicates the boundary of the i-th road; When modeling the environmental vehicle potential field model: to ensure the driving safety of the vehicle and maintain a safe driving distance as a constraint, the lateral and longitudinal potential field models of the environmental vehicle are constructed, wherein the safety distance of the environmental vehicle potential field model depends on the driving speed of the environmental vehicle potential field model. The faster the driving speed, the larger the safety distance is set; the slower the driving speed, the smaller the safety distance is set; The vehicle driving environment potential field model is constructed by combining the road boundary potential field model, the lane line potential field model and the environmental vehicle potential field modeling; A trajectory optimization module is used to set the vehicle as a mass point, obtain the driving state of the vehicle based on the mass point, obtain the driving trajectory of the vehicle, input the driving trajectory of the vehicle into the improved traditional vehicle driving environment model, and output the optimized optimal driving trajectory, wherein the improved traditional vehicle driving environment model is solved based on a pre-established model predictive control and artificial potential field method vehicle trajectory planner, and the pre-established model predictive control and artificial potential field method vehicle trajectory planner is based on genetic algorithm for parameter optimization; The trajectory optimization module improves the traditional vehicle driving environment model and solves the vehicle trajectory planner based on the pre-established model predictive control and artificial potential field method, including: Generate safe operation trajectory according to vehicle motion status and design overtaking trajectory optimization strategy; Genetic algorithm is used to optimize the parameters of the vehicle trajectory planner of the pre-established model predictive control and artificial potential field method, and the optimized optimal driving trajectory is output; The optimal driving trajectory in the trajectory optimization module includes: road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation and lateral acceleration increment; The objective function of road centerline tracking in the trajectory optimization module is: The objective function of the environmental vehicle potential field is: The objective function of the potential field in the feasible overtaking area is: The objective function of the road boundary potential field is: The objective function of trajectory steepness evaluation is: The lateral acceleration increment objective function is: in 、 、 、 、 、 They are the target item weights of road centerline tracking, environmental vehicle potential field, overtaking feasible area potential field, road boundary potential field, trajectory steepness evaluation, and lateral acceleration increment. 、 、 、 are target values, represents the i-th road boundary, , For time, is the acceleration, is the trajectory steepness penalty, i.e., the difference between the vehicle position and the vehicle offset angle; The function formula to be solved is: represents the objective function, represents a linear inequality constraint, represents a linear equality constraint, and Represents upper and lower boundary constraints, represents a nonlinear constraint function.
5. The vehicle automatic driving trajectory planning system based on edge computing according to claim 4 is characterized in that: When the vehicle driving environment potential field model is improved by using the artificial potential field method in the improved model module, the road boundary, lane line and environment vehicle potential field models are respectively established. The road boundary, lane line and environment vehicle potential field models are used for the vehicle to drive along the center line when driving and not changing lanes, and it is forbidden to drive out of the boundary line; The vehicle trajectory planner of the model predictive control and artificial potential field method pre-established in the trajectory optimization module performs parameter optimization based on the genetic algorithm to obtain the process of the optimal driving trajectory, including: obtaining the optimal lateral acceleration by solving the improved traditional vehicle driving environment model, processing the optimal lateral acceleration discrete sequence by curve fitting, optimizing the model predictive control and the artificial potential field method by cross calculation, combining the model predictive control rolling optimization mechanism in the pre-established vehicle trajectory planner of the model predictive control and artificial potential field method with the potential field model of the artificial potential field method, and finally solving the vehicle trajectory planner to obtain the optimal solution, and the vehicle trajectory planner with the optimal solution solves the improved traditional vehicle driving environment model to obtain the optimal driving trajectory.
Citation Information
Patent Citations
Vehicle collision avoidance dynamic safety path planning method based on accurate trajectory prediction
CN112109704A
Algorithm for planning lane changing track of automatic driving vehicle based on potential energy field heuristic search
CN112622932A
Autonomous driving vehicle trajectory planning and tracking control method considering active safety
CN115743174A
Constraint-fused high-speed automatic overtaking motion planning and control method
CN116424326A
Intelligent vehicle dynamic environment trajectory planning system and method based on adaptive potential field
CN117111610A
