Intelligent Vehicles Lane-changing Intention Identification Method with Driving Style Recognition

For intelligent driving systems, predicting the lane change intentions of surrounding vehicles in advance is essential to improve safety and efficiency in dynamic traffic conditions. In this paper, a lane-changing intention identification method with driving style recognition is proposed to identify lane-changing intentions for intelligent vehicles, incorporating driving style recognition to enhance prediction accuracy. Firstly, a dynamic clustering framework integrating the Gaussian Mixture Model is introduced to identify the driving style of vehicles under different traffic conditions. Subsequently, a lane-changing intention recognition model based on bidirectional long short-term memory networks is proposed. By leveraging driving style enhancements, the model is refined to better simulate and comprehend driving behaviors on the road. Finally, the NGSIM dataset is used to train and evaluate the proposed prediction identification method. The results show that the accuracy is improved by 2.1% compared to state-of-the-art methods.