Integrated Planning And Control For Autonomous Vehicle Collision Avoidance Systems

Vehicle collision avoidance system,which is an important implementation of vehicle safety,has become the research focus and difficulty in the field of autonomous driving technology.When an obstacle appears in front of the vehicle's front path,automatic braking or steering can help the vehicle avoid collision with obstacle,which significantly improves vehicle safety.There is still a big gap between the research and the practical application of the traditional collision avoidance system because of the variability of the driving environment of the autonomous vehicle and the limited working range of the vehicle actuator.Therefore,an integrated trajectory planning and motion control for autonomous vehicle collision avoidance system is investigated in this paper based on model predictive control.The main research contents of the thesis are as follows:Firstly,the structure of the traditional separated vehicle collision avoidance system and the related technical problems are analyzed.An autonomous vehicle collision avoidance system based on integrated trajectory planning and motion control is proposed.The system is based on model predictive control and the obstacles and road boundaries are considered in real time.This control method's effectiveness and scene customization are also analyzed.Secondly,the vehicle single track dynamic model and the magic tire model are established based on the Newtonian mechanics method.The steering and braking models are established according to the vehicle's collision avoidance maneuver,and the steady-state steering characteristics of the vehicle model are verified.These vehicle models are the basis of the integrated collision avoidance system design.Finally,the performance of the model predictive control method in the static obstacle scene is verified according to the pre-planned trajectory and the state information of the vehicle model.Thirdly,based on the coupled longitudinal and lateral dynamics,a nonlinear model predictive control scheme for integrated trajectory planning and vehicle dynamics control for obstacle avoidance maneuver is designed.The problem of nonlinear constrained optimization based on vehicle dynamics model is solved,and environmental information constraints such as obstacles and road boundaries and vehicle dynamics constraints are satisfied.In addition,the collision avoidance system uses terminal collision avoidance constraints to plan a safety trajectory under the existence of obstacles outside the predicted time domain.Finally,the typical scenarios of single and multiple obstacles are analyzed,and the effectiveness of the method in each scenario is verified in the simulation.Finally,due to the high computational time required by the solver,the implementation of nonlinear control problems in real-time environment is important.In order to solve the computational complexity problem of the nonlinear integrated scheme,a continuous linearization technique is proposed to transform the nonlinear problem into a convex quadratic problem that can be solved using a standard optimization solver.The robust tubebased model predictive control method is used to regard the linearization error as added uncertainty and the improved constraints are derived to ensure the stability of the uncertain system.Finally,the effectiveness and real-time performance of the rapid integrated planning and control method are verified in the simulation and tests.