Research On Lane Changing Control Of Intelligent Vehicle Based On Experienced Driver Steering Characteristics

Intelligent vehicle(IV)integrates environment perception,trajectory planning,and tracking control,replaces manual operation with automatic driving and makes up for the insufficiency of driver's ability.The uncertain factor of ‘human' from the road system composed of ‘human-vehicle-environment' can be removed,which can avoid traffic congestion and accident caused by human factors to the greatest extent.However,relevant research shows that the comfort of intelligent vehicle is far from that of skilled driver when driving automatically,especially in steering control.During the running of the vehicle,the lane changing operation is often performed due to obstacle avoidance,overtaking and traffic signal indication,and the longitudinal and lateral displacement and speed of the vehicle are changed.In this paper,intelligent vehicle lane changing control is taken as the research object,and the related research work of IV lane changing trajectory planning and trajectory tracking is carried out.The experiments of experienced driver's lane changing control characteristics acquisition based on real vehicle and driving simulator were carried out and the test data were analyzed.Test vehicles were added some sensors to record the vehicle test data.The longitude and latitude coordinates of the lane changing trajectory recorded by GPS were converted into two-dimensional plane coordinates in the geographic coordinate system.The test trajectory was analyzed,and the influence of intermediate position state constraints on the lane changing trajectory was studied.The hexagonal polynomial was chosen as the expression of lane changing trajectory.Aiming at the strong non-linearity of the experienced driver's lane changing control,an IV trajectory planning system was designed based on the neural network algorithm.The GA-BP neural network model was established.The influence of different factors on lane changing trajectory was analyzed and the experienced driver model was established.The model was trained and validated by real vehicle and driving simulator test data.The simulation results show that the trajectory generated by the lane changing trajectory planning system proposed in this paper is in good agreement with the trajectory selected by experienced drivers under the same conditions,and the error is small.The yaw rate of vehicle traveling along the specific trajectory in an unbiased form was analyzed.Based on this,the IV trajectory tracking yaw rate generator was designed.7-DOF vehicle dynamics model and Dugoff tire model were established.Based on the fuzzy sliding mode algorithm,the IV lane-changing trajectory tracking control system was established.A simulation platform was built with CarSim and Simulink and the trajectory tracking control simulation was carried out.The simulation results show that compared with the trajectory tracking control based on lateral deviation,the lane changing trajectory tracking control system based on yaw rate proposed in this paper has higher tracking accuracy and smaller yaw rate amplitude,which can improve vehicle comfort when tracking the ideal trajectory accurately.The hardware-in-the-loop simulation test platform of intelligent vehicle trajectory tracking based on D2 P rapid prototyping development platform,NI real-time simulator and host computer was built.The VeriStand software was used to compile and write the executable code of the vehicle model into NI real-time simulator.The trajectory tracking control algorithm was developed by using D2 P rapid running development platform.Based on the HiL simulation test platform,the trajectory tracking control experiment was carried out.The results show that the trajectory tracking results of the controller designed in this paper are consistent with that of the CarSim simulation,which verifies the stability and effectiveness of the controller.