Research On Multi-vehicle Cooperative Lane Changing Under The Condition Of Intelligent Network

In recent years,the automatic lane change technology of vehicles,which is one of the cores of driverless technology,has developed rapidly.With the continuous upgrading of V2 X technology,the popularity of vehicle sensor technology is increasing,and multi-vehicle collaborative driving technology is gaining more and more attention.The goal of vehicle collaborative driving is to efficiently use road conditions such as road test equipment while the vehicle is running.It is safe and efficient to achieve coordination between multiple vehicles.As a result,coordinated driving responds quickly,and it is also possible to complete coordinated operations such as acceleration,deceleration,and lane change operation according to actual conditions.However,there are few studies on multi-vehicle cooperative control,and there are few models used to describe the interaction of the system.The current research mainly focused on the system structure design of multi-vehicle cooperative driving based on intelligent networked vehicles.While there is relatively little research on the self-vehicle control algorithm and the surrounding vehicle cooperative control algorithm.The research scenarios mainly focused on the compulsory lane changes such as confluence areas,which cannot meet the complex and changeable traffic environment in the actual situation.Due to the high dimensions of collision avoidance constraints and the nonlinear nature of vehicle kinematics,optimal control problems are often difficult to solve.How to construct a centralized decision-making and distributed control multi-vehicle cooperative lane change system based on the characteristics of real-time communication of intelligent networked vehicles to meet the requirements of complex and dynamic changes in actual road conditions needs further solutions.This paper proposes a multi-vehicle cooperative safe lane change strategy under the condition of intelligent network connection.First,obtain the vehicle’s motion parameters and road traffic information through vehicle information perception system,such as vehicle’s speed(acceleration,accelerator,throttle opening,brake pressure,front wheel)through the vehicle’s sensors(GPS,radar,speed sensor,etc.),and state variables such as rotation angle and latitude and longitude position,environmental information are also needed.Then calculate data such as the angle between the vehicle and the lane centerline based on the map information.Secondly,through the establishment of the incentive model-based lane-change revenue function to make cooperative lane-change feasibility decisions.On the basis of meeting the safety requirements of the vehicle,determine whether vehicle’s cooperative lane-changing behavior is better than lane-keeping behavior under the current situation,which is used to make a decision judgment.Thirdly,based on model predictive control,a multi-objective optimized control function for cooperative lane change is established to achieve distributed control of the lane change process.A two-stage collaborative lane change framework is proposed,which divides the lane change process into a sparse longitudinal distance stage and a lane change stage.In the sparse longitudinal spacing stage,the lane change vehicle expands the longitudinal distance from the front and rear vehicles to achieve the safety distance required for lane change.Problems that are difficult to solve for optimal control functions caused by kinematic nonlinearity can be solved.Then,the rolling optimization time domain algorithm is used to solve the optimization control problem gradually and dynamically.Finally,through Matlab / Carsim joint simulation,the superiority of the proposed cooperative lane changing strategy in improving the feasibility of lane changing and driving comfort is verified.The feasibility and accuracy of this strategy are verified based on the scenarios in the US NGSIM open source traffic flow database.