Intention-Aware Dual Attention Based Network for Vehicle Trajectory Prediction

Accurate surrounding vehicle motion prediction is critical for enabling safe, high quality autonomous driving decision-making and motion planning. Aiming at the problem that the current deep learning-based trajectory prediction methods are not accurate and effective for extracting the interaction between vehicles and the road environment information, we design a target vehicle intention-aware dual attention network (IDAN), which establishes a multi-task learning framework combining intention network and trajectory prediction network, imposing dual constraints. The intention network generates an intention encoding representing the driver’s intention information. It inputs it into the attention module of the trajectory prediction network to assist the trajectory prediction network to achieve better prediction accuracy. The attention module in the trajectory prediction network mainly includes spatial attention module and channel attention module to reflect the relative importance of the influence of neighboring vehicles on the target vehicle. The attention module not only reflects the correlation between the target vehicle and the neighboring vehicles in the spatial position but also indicates the correlation between the target vehicle and the neighboring vehicles in the high-dimensional feature of the channel. We conduct ablation experiments on the NGSIM dataset to demonstrate the facilitation effect of our proposed intention network on the trajectory prediction task and the adjustment effect of the dual attention mechanism on the influence weights of neighboring vehicles. We also compare our model with some state-of-the-art models, and experimental data show that our network outperforms the current state-of-the-art on the publicly available NGSIM dataset.