基于时间序列的目标车辆轨迹预测算法

针对高速道路场景,对智能车辆前方的目标车辆进行轨迹预测。根据车辆运动轨迹数据具有时序性的特点,并为了增加轨迹特征的表征能力和上下文时空关联性,提出了将车道线特征、目标车辆的特征与历史轨迹数据的特征进行融合,和LSTM-CNN-LSTM融合模型,以提高目标车辆轨迹预测的精度。

Target Vehicle Trajectory Prediction Algorithm based on Time Series

In this paper,aiming at the high-speed road scene,the trajectory of the target vehicle in front of the intelligent vehicle is predicted.According to the temporal characteristics of vehicle motion track data,and in order to increase the representation ability of track characteristics and context spatiotemporal correlation,this paper proposes to integrate lane line characteristics,target vehicle characteristics and historical track data characteristics,and the lstm-cnn-lstm fusion model,in order to improve the accuracy of target vehicle trajectory prediction.