Overtaking Control Methods For Autonomous Vehicles

Overtaking is one of the most common driving behaviors as well as one of the easiest behaviors to trigger accidents or traffic congestion.The research on overtaking control methods for autonomous vehicles aims to enable the autonomous overtaking to be more effective,rational,secure and comfortable than the operation of drivers.Following researches have been carried out on decision making,trajectory planning and tracking control in the paper:Problems with decision making for cooperative overtaking of multiple vehicles have been solved through the research on the cooperation strategy based on vehicle information communication.Distance and speed of relevant vehicles during the overtaking have been analyzed to conduct overtaking decision making among relevant vehicles by satisfying collision prevention conditions and combining self-organized internetwork communication.The aim to achieve safe and effective overtaking has been reached through collaborative overtaking behaviors.Khepera experiment platform has been adopted to verify the effectiveness of the collaborative overtaking strategy.Through the introduction of Bezier curve locus,problems with trajectory planning in the overtaking have been solved.In comprehensive consideration with vehicles,under the premise of being restricted by safety,comfort and traffic efficiency in the traffic environment and in combination of third-order Bezier curve,an optimal overtaking trajectory by changing lanes has been figured out which can be updated in real time as well,to complete the action of overtaking.Problems with trajectory tracking control for two-wheeled vehicles have been studied.Kinematic models have been established for two-wheeled vehicles with more functions and better controllability.The feedback controller based on Lyapunov direct method has been designed to realize the trajectory tracking control of vehicles.Matlab Software has been adopted to conduct simulation experiments separately on the straight trajectory and arc trajectory of vehicles.Moreover,the validity and reliability of the controller have been verified through the experiment platform.