In this paper,the vehicle trajectory data is filtered and Gaussian smoothed,and different driving behaviors are classified and driving intentions are labeled.Subsequently,this paper proposes an LSTM-based vehicle lane change intention recognition model,which fully considers the interaction between vehicles,effectively extracts the temporal continuous features in the lane change process,and captures the local and long-term dependencies in the vehicle trajectory.The model takes the driving data of the interaction information between the target vehicle and its surrounding vehicles as input.Experimental results show that the proposed model has an accuracy of 92.85% in predicting vehicle lane change intention,and is significantly better than other models in a variety of evaluation indicators,showing the application potential in the actual traffic environment.
关键词
换道意图识别/机器学习/LSTM模型/智能交通
Key words
Lane Change Intention Recognition/Machine Learning/LSTM Model/Intelligent Transportation
维普
万方数据