Research On Path Planning Method With Obstacle Avoidance For Intelligent Vehicle

With the development of society and the advancement of science and technology,intellectualization has become a main developing direction of vehicles.The intelligent vehicle is a combination of technologies involving in computer,sensor,communication,artificial intelligence and automatic control.Its core includes three modules: environment perception,path planning and vehicle motion control.For the intelligent vehicle,a path planning method under global static environment and local dynamic environment is studied in this thesis,the final path results are obtained,and the obtained path is tracked.Firstly,global static path planning for intelligent vehicle based on Particle Swarm Optimization(PSO)is studied.To meet the real-time performance and safety requirement of vehicle path planning,a method combined by nonlinear inertia weight and linear learning factors is proposed to improve the PSO algorithm.An obstacle avoidance strategy that contrasts fitness values is adopted to process the collision problem between the paths and the obstacles,thus global path searching is completed.Compared to the traditional PSO,the improved PSO proposed in this thesis can obtain shorter path,better real-time performance and higher planning success rate.Secondly,in order to avoid obstacles safely under dynamic environment,an improved Artificial Potential Field(APF)is applied to local dynamic path planning for intelligent vehicle.The problems are solved by discretizing the boundary of obstacles to ensure the safety of avoidance,adding random escape force to escape the local minimum and considering the speed and acceleration of obstacles to apply traditional APF to dynamic path planning.The simulated results show that the improved APF can obtain local obstacle avoidance paths with better safety and real-time performance.Finally,the tracking of above planned paths is carried out by adopting joint simulation technology.The kinematics model of the vehicle is established.According to the lateral deviation and angle deviation,a yaw rate fuzzy controller of the vehicle is designed.The planned path is simulated by using ADAMS / Simulink.The simulation results show that the vehicle model can track target path effectively.This thesis aims to study the path planning method for intelligent vehicle,and the conclusions obtained can provide a reference to actual application of intelligent vehicle.