Research On Obstacle Avoidance Path Planning Of Intelligent Vehicle

With the development of China’s national economy,more and more vehicles are owned,and all kinds of traffic problems follow at the same time.In view of the main problem of frequent traffic accidents,in addition to strengthening the protection of people by passive safety technologies such as seatbelts and airbags,major vehicle enterprises have also increased efforts to develop Advanced Driving Assistant System to actively intervene in vehicle driving to avoid or mitigate collisions,such as automatic emergency braking system,lane departure alarm system,etc.However,sometimes these systems can only reduce the degree of injuries and can not completely avoid collision.In this paper,the path planning and tracking of intelligent vehicles to avoid obstacles are studied,which can guide vehicles to stay away from obstacles,fundamentally solve the occurrence of traffic accidents,and help people to drive safely.Because the artificial potential field algorithm has the advantages of small computation and high real-time performance,it is selected as the main method of trajectory planning in this paper.Aiming at the problem that the target point is not reachable and the local optimum exists,the traditional artificial potential field algorithm is improved by changing the repulsion potential field function and rotating the repulsion force at a certain angle.Then the road boundary repulsion potential field and the speed repulsion potential field are established to avoid obstacles in the current dynamic environment.Finally,a model predictive controller is established to track the trajectory of the improved algorithm,which verifies the effectiveness of the algorithm.The main research contents are as follows:(1)Establish vehicle dynamics model,including three degrees of freedom: transverse,longitudinal and yaw,which provides a prediction model for trajectory tracking module.Because of the high requirement of real-time in trajectory tracking,the nonlinear dynamic model can not meet the requirements,so the nonlinear dynamic model is linearized,which lays the foundation for the follow-up model predictive tracking control.(2)In view of the problem that the target point of is not reachable and the local optimal in the traditional artificial potential field method,this paper improves the traditional artificial potential field algorithm by changing the repulsion potential field function and rotating the repulsion force at a certain angle.In order to verify the effectiveness of the improved algorithm,the traditional artificial potential field algorithm and the improved algorithm is compared in MATLAB.(3)Because there are dynamic obstacles in the real traffic environment,the artificial potential field method is added with the speed element,and the road model is established to meet the dynamic collision avoidance requirements of the vehicle in the actual traffic environment.The trajectory planning simulation of the compound artificial potential field is carried out in MATLAB.The trajectory is smooth,which meets the requirements.(4)Based on the model prediction algorithm,the trajectory tracking controller is established.By building the SIMLINK/Car Sim joint simulation platform,the simulation of the intelligent vehicle following the trajectory under the condition of double line shifting is carried out.It is proved that the controller can ensure the vehicle track the trajectory stably and safely.Finally,by setting different driving speeds and different road adhesion conditions,the tracking effect of the controller on the trajectory planned by the artificial potential field method is simulated: when the road conditions are good,the vehicle can track the trajectory smoothly at different speeds;When the road conditions are poor,the low-speed vehicle tracking trajectory works well(generally,when the road conditions are poor,the vehicle usually runs at low speed).The simulation results further verify the rationality and effectiveness of the improved artificial potential field method.In this paper,the improved artificial potential field method is used for trajectory planning.In order to further verify the effectiveness and rationality of the planned trajectory,the model predictive control is used to establish a trajectory tracking controller,which can better track the trajectory.This paper provides a reference for the research of intelligent vehicle trajectory planning.