The highway autonomous lane changing system is an important part of intelligent vehicles.It can sense the vehicle’s driving environment in real time through sensors,analyze the current driving risk autonomously,dynamically plan the driving trajectory and track tracking control.It has the great advantages of reducing driving fatigue and improving road utilization.huge potential.Traditional vehicle autonomous lane changing simulation research mostly presupposes ideal road curvature and fixed vehicle-vehicle interaction motion,and directly transmits data to the controller by matching the sampling period,ignoring the process of intelligent vehicle sensor perception and fusion.However,the real traffic scene involves multi-dimensional information such as lane boundaries,interactive vehicles,and pedestrians,and presents strong coupling characteristics as a whole.Traditional methods are bound to be difficult to verify the reliability of the system in the real road environment.Therefore,this paper takes the autonomous lane-changing system of intelligent vehicles as the research object,and mainly solves the problem of target vehicle identification and autonomous lane-changing control under heterogeneous sensor information.The specific research contents are as follows:Research on autonomous lane change decision-making mechanism and trajectory planning.By studying the scene characteristics of high-speed sections and drivers’ lane changing behavior,an autonomous lane changing intention identification and lane changing behavior execution strategy based on space and time expectation satisfaction was constructed.According to the typical traffic accidents caused by lane changing on the expressway,the safe lane changing distance model between this lane and the target lane and the quintic polynomial lane changing trajectory planning model were constructed,and the autonomous lane changing trajectory constraint and its objective function were introduced,which could solve the optimal autonomous lane changing trajectory on the premise of considering the lane changing safety and efficiency,so as to provide a theoretical basis for the lane changing trajectory tracking control of subsequent intelligent vehicles in high-speed sections.Research on autonomous lane change trajectory tracking control.A three-degree-of-freedom vehicle dynamic model was constructed,and a lane-changing trajectory tracking controller based on adaptive MPC control was designed.In the control process,in order to reduce the complexity and difficulty of nonlinear system in MPC algorithm control and improve the real-time performance of system control,the approximate linearization method was used to linearize its model,and a relaxation factor was introduced into the objective function to ensure control.The controller had a certain feasible solution capability at each sampling time.A polynomial lane-changing trajectory tracking simulation scenario was constructed in Matlab/Simulink software to verify the performance of the controllerResearch on heterogeneous multi-sensor information fusion strategy.By studying the current mainstream sensors of intelligent vehicles and their arrangement schemes,combined with the parameters required by the controller for autonomous lane-changing control,the arrangement scheme of sensors on high-speed sections of intelligent vehicles was designed.A heterogeneous multi-sensor information fusion strategy was constructed,and the effectiveness of the heterogeneous multi-sensor information fusion strategy in this paper was verified by carrying out high-speed road section simulation in the Matlab/Simulink autonomous driving module.Simulation research of autonomous lane changing system based on heterogeneous sensor information.Relying on Matlab / Simulink software,the scene of target vehicle low speed,target vehicle stationary and target lane vehicle cutting in front of the lane were constructed.By introducing the sensor module and setting the working parameters of the sensor,the simulation of the autonomous lane changing control system under the heterogeneous sensing information was carried out. |