Multi-lane Cooperative Formation And Control Of Intelligent Connected Vehicle Based On Virtual Leaders

With the rapid development and wide application of 5G communication,mobile internet,artificial intelligence and other technologies,self-driving technology has become one of the research hotspots in the global scientific and technological community.The problem of road traffic congestion is becoming more and more serious due to the increase of car ownership.Only increasing the supply of road facilities to alleviate the traffic pressure encounters a bottleneck,especially the planned land-use for roads in large and medium-sized cities has reached the limit.Therefore,improving traffic efficiency through necessary traffic operation control should become the main means to solve the traffic problem at this stage.The cooperative formation driving of intelligent connected vehicles is an effective way to reduce environmental pollution,improve road traffic operation,alleviate traffic congestion and improve road driving efficiency.However,due to the asymptotic development of technology promotion,the future road traffic will inevitably experience the mixed traffic stage of self-driving vehicles and human pilot vehicle sharing the road.Therefore,it is of great significance to study the group characteristics of ICV with high intelligence and connected under mixed traffic mode,especially the road traffic efficiency improvement potential brought by group cooperative control.Therefore,this paper studies the cooperative formation,formation optimal trajectory planning and tracking control of ICV on expressway driving.Firstly,the architecture of ICV cooperative formation control system based on virtual leaders is constructed,the control law of virtual leader is designed,the tracking behaviour and synchronization behaviour of individuals are introduced.A four-component model is established by multi-agent system theory to study cooperative formation control,and an ICV multi-lane cooperative formation control model based on virtual leaders is proposed to extend multi-ICV cooperative formation from one-dimensional queue to two-dimensional formation.The distributed model predictive control algorithm based on virtual leaders is designed to analyze the internal stability and string stability of ICV formation system.The simulation results show that the ICV formation system using distributed model predictive controller has better internal stability and string stability.Secondly,three key problems in ICV formation planning layer are studied,and the ICV formation evaluation indexes are proposed from safety and efficiency.The target node of ICV formation is the virtual leader,and the diamond is used as the basic formation.The artificial potential field method is used to model the influencing factors of road environment and the basic virtual leader.On this basis,the zero point of resultant force is used as the real-time positional posture of the virtual leader to get real-time target formation.The sales contract workflow is used to describe the pairing between ICV and virtual leader in the formation.Taking the optimal energy efficiency of the whole formation system as the objective function,the optimal task assignment matrix in the sampling period is obtained.The ICV and virtual leader corresponding to the intersecting trajectory are divided into subgraphs for analysis.The priority of ICV in each subgraph is obtained by using the sequential multi-vehicle trajectory decision algorithm.The low priority regards the high priority as a moving obstacle and avoids it.The traffic efficiency of HPV and ICV under different vehicle density and ICV permeability is simulated.The simulation results show that when the expected speeds are 70km/h,90km/h and 110km/h,respectively,compared with the traditional driving scenario,the maximum traffic efficiency of ICV formation with cooperative formation control strategy is improved by 47.1%,35.7%and 31.8%,respectively.Thirdly,the optimal trajectory planning problem of ICV is solved by radau pseudospectral method and interior point method.The hybrid A^*algorithm is used to obtain the waypoints of the individual ICV.Taking the connection of the waypoints as the reference line,the solution interval of the planned trajectory is limited in the neighborhood of the reference line.The avoidance of static obstacles is realized by adding soft and hard constraints,and the avoidance of dynamic obstacles is realized by speed replanning.A simulation environment based on AMPL(a mathematical programming language)and MATLAB is built.The simulation results show that the proposed ICV optimal trajectory planning algorithm is effective and has good real-time performance.Fourthly,the complex ICV trajectory tracking problem is decomposed into longitudinal motion control and lateral trajectory tracking control.The ICV longitudinal motion control is described by the state equation of the first-order inertial system,and the controller is designed based on the model predictive control(MPC)algorithm.The ICV lateral trajectory tracking control is described by the two degree of freedom vehicle dynamics model in the body coordinate system,and the controller is designed based on the linear time varying model predictive control algorithm.The tracking effect is tested by using Car Sim/Simulink simulation platform.The results show that the longitudinal motion control and lateral trajectory tracking effect of ICV are good.Finally,the simulation tests and actual vehicle verification of single ICV formation trajectory algorithm,and the simulation tests of multi-ICV cooperative formation control strategy are carried out.A Car Sim/Simulink simulation platform is built to realize the simulation of formation trajectory based on sampling period.Using the self-driving platform of intelligent electric vehicle S50EV-idea,the multi-scene real vehicle tests of single ICV trajectory planning and tracking performance are carried out.The results show that the intelligent electric vehicle trajectory is highly correlated with the single ICV simulation trajectory.The typical formation scenarios simulation are carried out by using the multi-ICV formation control simulation platform based on Pre Scan/Simulink.The results show that the multi-ICV cooperative formation control strategies are effective.To sum up,this paper analyzes the stability of the ICV formation system driving on expressway,studies the relevant algorithms of cooperative formation,and puts forward the multi-lane ICV formation trajectory planning method and formation trajectory control method.Thus,this study can provide a theoretical basis for the research and development of ICV cooperative formation control technology.