MALS-Net: A Multi-Head Attention-Based LSTM Sequence-to-Sequence Network for Socio-Temporal Interaction Modelling and Trajectory Prediction

Predicting the trajectories of surrounding vehicles is an essential task in autonomous driving, especially in a highway setting, where minor deviations in motion can cause serious road accidents. The future trajectory prediction is often not only based on historical trajectories but also on a representation of the interaction between neighbouring vehicles. Current state-of-the-art methods have extensively utilized RNNs, CNNs and GNNs to model this interaction and predict future trajectories, relying on a very popular dataset known as NGSIM, which, however, has been criticized for being noisy and prone to overfitting issues. Moreover, transformers, which gained popularity from their benchmark performance in various NLP tasks, have hardly been explored in this problem, presumably due to the accumulative errors in their autoregressive decoding nature of time-series forecasting. Therefore, we propose MALS-Net, a Multi-Head Attention-based LSTM Sequence-to-Sequence model that makes use of the transformer's mechanism without suffering from accumulative errors by utilizing an attention-based LSTM encoder-decoder architecture. The proposed model was then evaluated in BLVD, a more practical dataset without the overfitting issue of NGSIM. Compared to other relevant approaches, our model exhibits state-of-the-art performance for both short and long-term prediction.

Automated summary

Predicting the trajectories of surrounding vehicles is crucial in autonomous driving, especially in highway settings where minor deviations can lead to serious accidents. Current methods use RNNs, CNNs, and GNNs to model vehicle interactions and predict future trajectories. However, popular datasets like NGSIM are criticized for being noisy and prone to overfitting. Transformers, known for their performance in NLP tasks, have not been extensively explored due to cumulative errors in time-series forecasting. To address this, we propose MALS-Net, a Multi-Head Attention-based LSTM Sequence-to-Sequence model that effectively utilizes the transformer's mechanism without suffering from cumulative errors. Our model outperforms other approaches in both short and long-term prediction on the practical dataset BLVD.