Research On Autonomous Lane Change Of Autonomous Vehicles

Lane change is one of the driving behaviors commonly used by drivers to operate vehicles.Due to driver’s inattention or operation error,it is the main cause of traffic accidents and road jams.Therefore,researching the autonomous lane changing assistance system of the vehicle,including the functions of lane changing decision,lane changing trajectory planning and tracking control,will help reduce traffic accidents,improve passenger comfort,and improve vehicle driving efficiency.Vehicle lane change can be divided into free lane change and forced lane change.Among them,forced lane change is mostly used at ramp junctions and accidents in front,the vehicle must perform lane change operations.While free lane change is more flexible and less researched.Therefore,this paper takes autonomous vehicles as the research object,mainly studies the free lane changing of vehicles,and establishes a safe and effective autonomous lane changing algorithm.Firstly,this paper analyzes the minimum following safety distance of the vehicle,and then proposes a vehicle lane change safety distance model.The simplified vehicle model of autonomous lane change behavior decision algorithm is established.The behavior decision problem is regarded as an optimization problem.Construct its logical constraints and drive at the optimal speed,choose the appropriate lane and the shortest travel time as the objective function.Finally,a decision-making method based on a mixed logic dynamic system is established,and a mixed linear integer programming method is used to optimize the solution.Secondly,based on the idea of model predictive control,an optimization method-based lane change trajectory planning method is established.The kinematics model is determined as a prediction model for trajectory planning and the lane change time is determined according to the motion relationship between vehicles.The objective function related to lane change time and vehicle steering performance is constructed and optimized.In view of the uncertainty of the environment vehicle,a logical decision to replan the lane change trajectory of the vehicle is determined to prevent the collision of vehicles along the original lane change trajectory due to changes in environmental vehicles,further ensuring the safety of the vehicle.Then a tracking controller based on the model predictive control for the lateral and longitudinal joint control is established.Construct a nonlinear prediction model for lateral and longitudinal joint control,and linear and discrete,according to vehicle actuator performance and lane changing environment to establish constraints,and establish the objective function of the lateral and longitudinal joint controller.Finally,the objective function is transformed into the form of quadratic programming for optimal solution.Finally,in order to test the performance of the autonomous lane changing algorithm designed in this paper,a simulation platform was built by Simulink,CarSim and PreScan software.Five kinds of operating conditions were designed for simulation verification.The simulation results show the effectiveness and robustness of the autonomous lane changing algorithm designed.