Study On Lane-changing And Overtaking Control Method Of Autonomous Vehicle

The goal of the research is the trajectory planning and tracking method of high-speed autonomous vehicle during lane changing and overtaking at high speed.Through the rapid control prototype simulation system to control the brushless direct current motor and to study the performance of the autonomous vehicle steering actuator.The research of planning layer is one of the key issues in the field of autonomous vehicle research,which is of great significance to the in-depth study of the problem.This paper presents a new path planning method based on the physical characteristics of vehicle motion.Then a trajectory tracking control method using model predictive control is developed,the method is validated by simulation for the high-speed working conditions.In addition,the effectiveness of the steering actuator control method is validated by the experimental platform.In order to study the vehicle handling stability,the 7 DOF vehicle dynamics model and magic tire model is adopted as the prediction model.On the one hand the dynamic characteristics can be described accurately to satisfy the requirement of the vehicle driving control.On the other hand,the control algorithm can reduce the computational burden under the special condition and satisfy the real-time requirements of operation.Using an optimized,four times Bezier curve to design a trajectory for vehicle with high-speed motion.The method can meet the requirement of the geometric constraint for collision avoidance,satisfy the constraints of continuous curvature and reduce the computation in the planning process.Associated with vehicle dynamic characteristics,this method set the path optimization index to satisfy the demands of different working conditions.Based on the nonlinear model predictive control method and considering the nonlinear characteristics of vehicle running at high speed,improved the model mismatch problem.This method can reduce the computation and improve the solution efficiency by transforming the nonlinear model into linear time-varying model and into a quadratic programming problem.The method adopts reasonable restrictions to meet other constraints of vehicles in real environment and enhance the stability of the method.This paper studies the principle of position control and speed control of BLDC,which is a steering actuator of the autonomous vehicle.In addition,through constructing the semi-physical simulation platform based on dSPACE to study the characteristics of the steering mechanism to make it meet the basic requirements of autonomous vehicle steering system.