Driving state evaluation of intelligent connected vehicles based on centralized multi-source vehicle road collaborative information fusion

When the intelligent vehicle is driving on the road, the driving state of the vehicle needs to be monitored and evaluated in real time, such as whether the distance from the vehicle or obstacle in front of it is often less than or close to the safe driving distance, whether it often deviates from the lane at high speed, whether the temperature of the power battery is often higher or lower, etc. The intelligent vehicle has rich environment perception systems and communication systems. A large amount of effective driving state information is obtained through these systems, then the information is fused by intelligent algorithm, which is very useful to evaluate the driving state. By using the developed Intelligent Connected Vehicle (ICV) information collection system platform, including vision, radar, environmental perception, positioning, and V2X communication system for obtaining roadside information, etc. various driving state parameters of the vehicle during the current driving process of the intelligent vehicle are obtained in real time when the vehicle is driving, then transmitted to the monitoring platform, and the intelligent immune algorithm is applied to establish model on monitoring platform. In the model, the state space representation of related problems is designed. The user establishes the autologous library according to the typical vehicle driving state in the monitoring road driving characteristics, which is used as the known antigen. The parameters of the real-time driving state of the vehicle are used as the unknown antigen, and the immune algorithm model is applied to get the evaluation of the current vehicle driving state and feedback the evaluation results to the vehicle and give early warning. In this paper, the model is verified by experiments, and the data are analyzed. In the experiment, the evaluation results of the monitoring platform are consistent with the state of the driving vehicle, which shows the feasibility of this model for the evaluation of the driving state of the ICV.