Research On Vehicle Obstacle Avoidance Path Planning And Tracking Control Based On Improved Repulsive Force Model

In recent years,with the development of computer and the Internet,the era of intelligence is changing the world quietly.As a widely used transportation tool,the development of car has been concerned.Autonomous driving technology is the trend of future automobile development and has become the research object of many universities and enterprises.Autonomous driving technology is mainly divided into perception positioning,planning decision and path tracking.As the main core technologies of automatic driving,the development of planning decision and path tracking plays an important role in the further promotion and application of automatic driving technology.Based on path planning and path tracking control,this paper studies the following contents:(1)Starting from local path planning,this paper makes improvements based on the defects of traditional artificial potential field,which is easy to fall into the unreachable target and local optimum.Firstly,the deflection direction of the repulsion force is determined according to the distance between the position of the vehicle and the road boundary,the position of the vehicle and the relative position of the speed direction and the obstacle,and then an angle is deflected in the deflection direction,so as to overcome the problem of local optimum.The repulsion function is redesigned,and the smoothness of the planned path is improved by adding the horizontal and longitudinal regulating factors,and the distance regulating factors are added to overcome the unreachable target problem.(2)According to the actual driving environment,the road boundary repulsive force field and the repulsive force range of elliptic obstacles are established,and the short and short axis of the influence range of elliptic repulsive force of obstacles is determined according to the minimum longitudinal steering safety distance of vehicles and the width of obstacles.Considering the velocity repulsive force field under dynamic obstacles and adding the vehicle dynamics constraint to the process of vehicle path planning,the planned path is more in line with the actual driving environment and higher trackability.(3)Based on the model predictive control theory,the three-degree-of-freedom dynamics model of the vehicle was taken as the predictive model of the model control,and the position relationship between the vehicle and the reference trajectory point was analyzed,and the lateral deviation is reasonably approximated.The equivalent control quantity boundary constraints are derived by considering the vertical and horizontal tire force coupling effect.Taking path accuracy and vehicle ride comfort as driving objectives,lateral deviation,yaw deviation and front wheel angle increment are designed as cost functions,which are transformed into constrained quadratic programming problems to calculate front wheel angle in real time.(4)Combining the artificial potential field path planning algorithm of the improved repulsive model and the model predictive control algorithm,the real-time planning control algorithm is established to cope with the changing driving environment.Finally,the Matlab and Carsim co-simulation platform is established,and the common static obstacle and dynamic obstacle avoidance scenes of automobiles are selected for simulation verification.The results show that the real-time dynamic programming control algorithm established in this paper can successfully realize the active obstacle avoidance function of vehicles,and the vehicles have high safety and stability in the process of active obstacle avoidance.The feasibility of the proposed algorithm is verified,which provides a high reference value for the path planning and tracking control technology of autonomous vehicles.