Decision-making And Control Method For Connected Autonomous Vehicle Based On Game Theory

With the continuous development of my country’s automobile industry,networking and intelligence have become the key development directions for future automobiles.At present,many domestic companies and universities are investing a lot of energy in related technologies for intelligent networked automobiles,in order to promote "no accidents and no jams." Intelligent transportation system is formed to avoid traffic accidents caused by improper decision-making and poor driving habits of drivers.Lane-changing behavior of vehicles is one of the most common behaviors and one of the behaviors that is most likely to cause safety accidents and traffic jams.Therefore,this article uses the intelligent networked vehicle as the carrier to carry out the study of lane changing behavior for expressways to improve the driving efficiency of vehicles and reduce the occurrence of traffic accidents caused by lane changing.Firstly,based on the idea of game theory,and uses the complete information non-cooperative dynamic game method to propose a vehicle lane-changing decision model based on game theory.First,judge whether there is a lane change motivation based on the perceived environmental information,and then coordinate with the surrounding vehicles to continue the feasibility judgment of the lane change.Finally,according to the characteristics of the lane-changing vehicles,the calculation methods of the safety benefit function,time benefit function and demand benefit function are proposed,and the Nash equilibrium definition is used The optimal strategy combination is solved for the vehicles in the game state.Secondly,through the analysis and comparison of the currently commonly used lane-changing trajectory planning methods,and the trajectory planning method to the fifth-order polynomial and the longitudinal fourth-order polynomial is proposed to improve the efficiency of trajectory calculation.The optimization of the lane-changing parameters and adjustment parameters in the trajectory is studied,and the constraint conditions in the lane-changing process are proposed.Using simulation software to carry out the lane-changing trajectory simulation experiment,after analyzing the multi-designed lane-changing trajectory in this paper,it meets the feasibility and stability requirements.Then based on the model predictive control algorithm to build a trajectory tracking controller for the horizontal and vertical joint control,while ensuring the smoothness of the tracking control under high-speed driving,and the system can realize to control requirements of the lane-changing vehicle at high speed.Finally,a joint simulation platform was built based on Prescan,and different working conditions were set to verify the effectiveness of the lane-changing decision module under different lane-changing motivations,the rationality of the revenue function,and the reliability and stability of the trajectory tracking control system.By comparing the results based on NGSIM,it is verified that the lane change decision model proposed in this article is better accuracy.