Trajectory Prediction Based on Roadside Millimeter Wave Radar and Video Fusion

In recent years, with the development of intelligent transportation system, roadside service is more and more widely used in urban road transportation system. Based on the roadside millimeter wave radar and camera, this paper obtains the motion trajectory data of vehicles close to the intersection, filters and processes the radar data, detects and tracks the video data, and then carries out data fusion to obtain more accurate trajectory data. Compared with a single sensor, the accuracy and stability of fused data are improved. In addition, the LSTM neural network is used to predict the vehicle trajectory and obtain the location information of the target vehicle passing through the intersection, so as to improve the traffic condition of the intersection.