On The Cooperative Adaptive Cruise Control With Lane Change Assistance

Vehicles with cooperative adaptive cruise control(CACC)can move forward together,which can not only reduce traffic congestion and collision accidents but also help to improve energy efficiency.Therefore,CACC has good development prospects.However,there are still a series of problems to be solved,such as how to ensure the string stability when the timestamps of all vehicles in the platoon are synchronized,how to reduce the adverse impact of human driving vehicles on the string stability of the hybrid platoon,how to make the connected vehicles coordinate with each other to complete the lane changing operation safely,how to realize the cooperative control when there are multiple communication topologies in the platoon.This paper has carried out systematic research on the above problems.A complete set of cooperative adaptive cruise control theories and methods are proposed,including longitudinal driving,lane change trajectory planning and collision avoidance,and adapting to a variety of communication topologies,which can realize the safe and efficient driving of the platoon under various restrictions such as timestamp synchronization,the existence of human drivers and obstacle interference.The research content of this subject mainly includes the following aspects:1)A distributed model predictive control algorithm with two-stage constraints to solve the problem of how to ensure the predecessor-follower(PF)string stability of the heterogeneous platoon when the timestamps of all vehicles are synchronized.Each vehicle has an independent computing unit in the control algorithm,which can use the driving information of other vehicles obtained in the previous sampling period to calculate the optimal control quantity at the current sampling time.The cost function of each vehicle's local optimal control problem is designed with tracking as the control objective,and its asymptotic stability is guaranteed by the terminal constraint method.To satisfy the conditional constraints of timestamp synchronization,we construct a two-stage constraint method to achieve PF string stability of the platoon.2)Considering the influence of human driving vehicles on the string stability of the hybrid platoon,a cooperative adaptive cruise control algorithm which can ensure the leader-follower(LF)string stability of the hybrid platoon is designed.The hybrid driving of human-driven vehicles and autonomous vehicles is a typical road condition in a short time.When the hybrid platoon encounters interference,to limit the impact caused by interference to a specific range,the hybrid platoon is required to meet the LF string stability.Therefore,aiming at the adverse effects of human-driven vehicles on the string stability of the hybrid platoon,a method to realize the LF string stability of the hybrid platoon is proposed.Based on the method,a cooperative adaptive cruise control algorithm is designed for the platoon combined with model predictive control and proportional differential control.The algorithm can optimize the hybrid platoon's comfort,economy and traffic efficiency on the premise of ensuring the LF string stability.3)Taking the merging scene as the research object,considering how to coordinate the connected vehicles to safely complete the task of changing lanes,a trajectory planning and tracking control algorithm is proposed.The lane changing process involves the safety of multiple vehicles in the current lane and the target lane.This paper studies the coordination problem of connected vehicles in the lane merging scene,and designs the trajectory planning and tracking algorithms for the vehicle to change lanes and the vehicles in the target lane.In the part of trajectory planning,firstly,an optimal lane changing trajectory is planned for the vehicle waiting to change lanes by combining the polynomial method and the optimization method.Then the vehicle to be changed send its planned trajectory to the platoon in the target lane.Combined with the received information and their current motion state,the vehicles in the target lane designed their own desired trajectory using the polynomial law.The planned trajectory contains a critical collision point.To avoid the collision accident caused by tracking error at this point,this paper uses model predictive control to design a trajectory tracking algorithm with obstacle avoidance constraints for the vehicle waiting to change lanes.4)Aiming at the problem that the external information received by vehicles in different communication topologies may change,a distributed model predictive control algorithm which can adapt to multiple communication topologies is proposed in this paper.The external information that each following vehicle can receive in different communication topologies may change.At this time,the cooperative control algorithm for single communication topology is no longer applicable.To solve the problem that there may be a variety of communication topologies in the platoon,different cost functions are designed for each following vehicle according to whether it has a communication connection with the leading vehicle.For following vehicles without direct communication with the leading vehicle,a terminal average constraint function is introduced to ensure that their actual outputs can converge to their planned desired outputs in a limited time.In addition,by adjusting the weight coefficients of the cost function,the algorithm can realize the Lyapunov stability condition under certain conditions,thus ensuring the tracking performance of the algorithm.