Research On Motion Planning Algorithm For Autonomous Driving Vehicles

With the rapid development of science and technology,autonomous driving technology is coming into our life at a rapid pace.However,the current mainstream autonomous driving system still has problems such as complex system software structure,large amount of calculation of detection algorithm,poor robustness of planning algorithm,which have become the main reasons that hinder the popularization of unmanned driving technology.In this paper,a dynamic obstacle avoidance motion planning algorithm based on optimal control theory is proposed.Firstly,the logic structures of vehicle collision detection algorithm and lane keeping algorithm are improved as important constraints for solving vehicle motion planning problems.Secondly,according to the requirements of driving efficiency,energy consumption and ride comfort in the process of vehicle movement,three performance index functions are established to replace the seven trajectory evaluation functions in the traditional algorithm as the objective function of the optimal control algorithm.Finally,according to the established constraint conditions and performance index function,the vehicle kinematics model was regarded as a controlled system,and the motion planning problem in the process of vehicle lane changing and obstacle avoidance was transformed into an optimal control problem to solve,and the vehicle motion planning algorithm based on the optimal control theory was established.In this paper,the logic structure of vehicle driving constraint algorithm is improved greatly,which reduces the calculation amount of detection algorithm.The planning algorithm makes full use of the structural advantages of the optimal control principle,and compared with the traditional motion planning algorithm,it can output additional vehicle control data corresponding to the trajectory and conforming to the constraint conditions.The planning module is integrated with the function of the control module,and the vehicle control data of the corresponding vehicle trajectory can be guaranteed to meet the performance constraint conditions,so that there is no need to do approximate processing for unreasonable control data.The proposed motion planning algorithm has strong robustness and can effectively reduce the complexity of automatic driving software system.According to the simulation results,the motion planning algorithm constructed in this paper can well complete the motion planning task under a variety of complex road conditions.Compared with the RRT path planning algorithm widely adopted in the autonomous driving field,the motion trajectory planned by the optimal control algorithm also has obvious advantages in the technical indicators of driving efficiency,energy consumption and riding comfort.