Research On Coordinated Control Strategy Of Lane Changing And Overtaking For Connected Automated Vehicle In Multi-lane Environment

Lane changing and overtaking are widespread driving behaviors of vehicles on expressways.Unreasonable lane changing and overtaking behaviors may lead to road traffic congestion,while 75% of lane changing accidents are caused by drivers’ insufficient perception of the surrounding environment.With the application of sensor and communication technology in vehicles,connected and automated vehicle(CAV)can share road condition information with surrounding vehicles in real time.A large number of research results show that cooperative driving strategy in CAV environment can effectively improve the traffic efficiency.With the development of vehicle communication technology,the emergence of new topological network and the change of topological network over time bring great challenges to multi vehicle control.The multi-agent system(MAS)theory is a powerful tool to study distributed network control.Therefore,this paper takes the intelligent vehicles as the research object,designs the distributed controller with reference to the multi-agent theory,and studies the multi vehicle cooperative lane changing and overtaking strategy in the scene of the Internet of vehicles.Firstly,this paper takes the multi vehicle cluster as the main application scenario,aiming at the problem of multi vehicle cooperative lane changing,designs a vehicle cooperative control framework based on multi cluster system.The vehicle unit(OBU)is equipped with sensors to sense the environment,and can carry out vehicle to vehicle(V2V)communication through wireless network.On this basis,intelligent networked vehicles select neighbor vehicles,leader and virtual leader as the state evolution of control protocol through distributed clustering algorithm in communication layer;space control algorithm suitable for cluster space allocation is designed in decision layer to sol ve the problem of lane change conflict caused by multi vehicle lane change at the same time;cluster control protocol constraint based on leader follower is established in control layer The trajectory of intelligent networked vehicles is analyzed,and the theoretical proof is given.The simulation results show that the proposed algorithm and control protocol can achieve multi cluster convergence consistency and achieve safe and accurate lane change under the networking technology.Overtaking is a more complex driving behavior than lane changing.When vehicles need to go from the original lane to the overtaking lane and then return to the original lane,there will be lane changing conflicts under uncertain conditions.For this reason,this paper takes single vehicle overtaking single vehicle and vehicle overtaking in queue as application scenarios,and proposes a step-by-step overtaking cooperation strategy based on queue control for vehicle overtaking control.Firstly,overtaking vehicles select other vehicles that have potential threat to overtaking safety as neighbors,and establish self-organizing network for communication;the system adopts the control model based on formation error in longitudinal direction,and the control model based on relative Lane lateral distance error in transverse direction.In different stages of overtaking process,each intelligent networked vehicle cooperates with the controller in transverse and longitudinal direction Now the formation is dynamically adjusted in cooperation,an d the whole lane changing process is finally completed.The simulation results verify the effectiveness of the cooperative control strategy of single vehicle overtaking and single vehicle overtaking in queue.