Research On Lane Change Control System Of Intelligent Vehicle Based On Cloud Server

With the increase in the degree of networked smart vehicles and the emergence of new technologies such as 5G and V2 X,single vehicle intelligence can no longer meet the actual needs of higher-order autonomous driving and requires the organic combination of vehicle,cloud,road and human to co-ordinate and plan the autonomous driving behaviour of vehicles.This paper is based on the cloud server for intelligent vehicle lane change control,completing the communication and monitoring between the cloud and the vehicle,realising remote control of autonomous driving,which can handle the vehicle in case of emergency failure of the intelligent vehicle and remotely control the vehicle to change lane to the emergency lane.The main research of this paper is as follows:A lane change decision,trajectory planning and trajectory tracking system on the cloud server side is built.The cloud side uses Spring Boot as the framework and Web Socket as the communication protocol,and uses topic communication to obtain sensory information,and then outputs the control quantity after the lane change decision,trajectory planning and trajectory tracking system is calculated,and releases the corresponding control topic to the vehicle side,so as to complete the control of lane change in the cloud.The software and hardware architecture of the in-vehicle terminal is built.The combined navigation device INS-550 D is used as the environment sensing unit to collect the position information of the lane change vehicle and the surrounding vehicles,and send the information to the on-board control unit EPC-P30X6 IPC based on the ROS operation platform.On the other hand,the on-board side receives the control signals via topic subscription and sends them to the on-board control unit,where the underlying actuator completes the lane change.A cloud server-based algorithm for lane change decision,trajectory planning and trajectory tracking is designed.The speed difference accumulation degree is used as the motivation for lane change,and the minimum safe distance for lane change is used as the condition to decide whether the vehicle can change lane under the current working condition and send the decision result to the trajectory planning system.The lane change trajectory planning system selects the time based quintuple polynomial as the lane change trajectory,and carries out the trajectory planning design for the lane change with accidental interference from unimpeded vehicles,and optimises it with the penalty function improved particle swarm algorithm,and finally gets the optimal lane change trajectory and sends it to the trajectory tracking system.The designed lane change trajectory tracking system uses a combination of Pure Pursuit,Stanley and PID control algorithms to achieve trajectory tracking while considering the cloud control delay,which is contrasted with the model predictive control algorithm to select the appropriate control algorithm for cloud controlled lane change.The simulation of lane change control based on cloud server and real-vehicle verification test were completed.Simulation platform was built in Matlab/Simulink in conjunction with Carsim for simulation verification of lane change trajectory tracking control,while the decision making,trajectory planning and tracking modules of the cloud-based lane change control system were tested in real vehicles,and the results of both proved that: the designed cloud server-based intelligent vehicle lane change control system has certain real-time,accuracy and smoothness,and can meet the requirements of autonomous lane changing on good roads.