Research On Obstacle Avoidance Path Planning Of Unmanned Vehicles Based On Improved Artificial Potential Field Method

With the continuous improvement of science and technology,society's demand for unmanned vehicles is increasing.Local path planning is one of the key researches in the development of unmanned vehicles.Unmanned vehicles follow certain parameter evaluation indicators to autonomously plan an optimal or sub-optimal obstacle avoidance path connected from the starting point to the target position.The rationality of path planning will directly affect the completion of unmanned vehicle tasks.According to the characteristics of the road environment and the constraints of vehicle dynamics,the algorithm of local path planning will be more complicated and should be more restricted.This article first analyzes the current path planning algorithm and explains the advantages and disadvantages of the existing algorithm;then based on the traditional artificial potential field method to carry out research,through the analysis of the algorithm,the two major problems of the algorithm are found and classified Discussion and improvement of the reproduction;Finally,in view of the insufficient application of the artificial potential field method in actual roads,on the basis of the improved artificial potential field method,combined with road environmental constraints and vehicle dynamic constraints,a road artificial The potential field model is used to carry out the simulation experiment of obstacle avoidance path planning according to the established algorithm model.The main research contents are as follows:(1)Through reading a lot of literature,the current global path planning algorithms and local path planning algorithms are analyzed separately,the advantages and disadvantages of these algorithms are summarized,and the current optimization methods of these algorithms are explained.(2)In-depth study of the traditional artificial potential field algorithm,to explore the two major problems of the algorithm's target unreachable and local minimum Field function;on the problem of local target minimum value,using sub-target point increase management,through the analysis of different distribution of obstacles in the environment,establish different sub-target point addition schemes,and finally carry out experimental simulations separately The established road artificial potential fieldmethod laid the foundation.(3)Based on the improved artificial potential field method in the road constraints,the characteristics of the algorithm,road constraints,and vehicle dynamics constraints will be comprehensively considered,and a static road artificial potential field model is proposed;considering that the algorithm cannot set the obstacle avoidance distance Problem,this paper builds a safety ellipse in the model based on the characteristics of the vehicle driving on the road and the vehicle dynamics theory,making the algorithm more practical in obstacle avoidance.(4)Aiming at the shortcomings of static road artificial potential field method for avoiding dynamic obstacles,this paper analyzes the typical situation of vehicle collision in the environment,comprehensively considers the distance between vehicles and the collision of driving directions,and establishes a vehicle collision based on the safe ellipse theory The calculation method of the danger coefficient is introduced into the repulsive force function of the road artificial potential field method,and a dynamic road artificial potential field method is proposed.(5)Improved algorithms are written in MATLAB to complete the path planning and simulation of unmanned vehicles in static and dynamic environments,and to optimize the planned paths.Based on the MATLAB and CARSIM joint simulation platform for obstacle avoidance path planning in complex environments The simulation further verifies the scientificity of the path planned by the algorithm.Based on the traditional artificial potential field method,this paper conducts research on obstacle avoidance path planning for unmanned vehicles.Simulation results show that the optimized algorithm can effectively plan better path trajectories.The significance of providing a theoretical reference.