Research on Self-driving Vehicle Driving Status Monitoring and Trajectory Tracking Technology

Considering the research of driving state monitoring and trajectory tracking technology of autonomous vehicles in roads under low permeability conditions is an important technical means to alleviate traffic problems and ensure traffic safety, which is of great significance to improve the safety and robustness of autonomous driving. However, the traditional trajectory tracking algorithm has the problem of insufficient data accuracy and cannot effectively obtain the holographic traffic trajectory state. Accordingly, this paper proposes a forward vehicle state prediction algorithm based on the combination of the Auto-Regressice Moving Average (ARMA) model and Kalman filter to achieve accurate prediction of vehicle trajectory. At the same time, this paper also proposes a spatiotemporal fusion strategy based on the topological statistical distance of vehicle network data and roadside wide-area radar data and constructs a spatiotemporal correlation mechanism of the absolute position of vehicle GNSS and relative position information of millimeter wave radar detection to realize continuous and accurate location tracking service of vehicles and provide reliable data support for intelligent traffic control.