Research On Remote Driving Assistance Safety System Based On Edge Computing

With the rapid development of 5G mobile communication technology,remote driving control of net-connected vehicles has become a reality,which can be applied not only in military fields such as tactical reconnaissance and mine clearance and explosive removal,but also in civilian fields such as security patrols,remote dispatching of vehicles and driving in dangerous scenarios.Therefore,it is of great theoretical and practical application value to carry out research related to remote driving of network-connected vehicles.This thesis takes the remote driving auxiliary safety system based on edge computing as the research object,researches the remote control architecture of the whole vehicle of the network-linked vehicle,completes the development and implementation of the remote driving control system software and hardware;and carries out the research on the vehicle safety warning strategy for the driving safety problems in the process of remote driving.In this thesis,the remote driving system is divided into three parts: remote vehicle side,driver side and server,and the overall architecture of the remote driving system is designed and the hardware and software scheme of the system is realized through in-depth research on remote driving related technologies,namely: vehicle CAN bus technology,wireless communication technology and network communication technology,etc.Firstly,we set up a local area network through wireless communication equipment to realize the penetration of intranet and complete the requirements of off-site network connection between all parts of the system;secondly,we design the hardware data collection module of remote vehicle end/driver end based on vehicle CAN bus protocol to realize the collection of vehicle driving status data/cockpit control vehicle data;finally,we use network communication technology to realize the data communication between vehicle end,driver end and server,and transfer the vehicle end collection data to the server.Finally,use the network communication technology to realize the data communication between vehicle end,driving end and server,upload the vehicle driving data collected at the vehicle end to the driving end for display,and at the same time,send the control data collected at the driving end down to the remote vehicle end for execution to realize the basic communication requirements of remote driving,and upload the collected data from both ends to the server for data storage.Considering the real-time demand for vehicle driving data from remote driving vehicles,this thesis selects 5G network as the wireless network of the system.In order to improve the safety of remote driving vehicles on the road,this thesis investigates the remote driving assisted safety function.By introducing edge computing technology,an assisted driving safety warning model is designed in the MEC device.The model calculates vehicle safety thresholds based on the remote vehicle collecting vehicle driving status data(such as speed,acceleration,heading angle,steering wheel turning angle,vehicle side tilt angle and other sensor data),and monitors and warns about the driving status of the remote vehicle as well as the driving environment for danger.Finally,different vehicle driving scenarios are built in the vehicle simulation software Carsim/Trucksim,and joint simulation with Matlab&simulink is performed to test the function of the safety warning model designed in this thesis.