Highlights With the modified Swin Transformer, the MTPT precisely infers multimodal trajectories of vehicles considering surroundings. The attention map provides model explainability to improve user acceptance by revealing the intrinsic mechanism of the MTPT. The influencing factors agree with the human driving experience, which indicates the MTPT has learned how to predict. The accessible grid is a useful prior knowledge of drivable areas that boosts the training process.
Abstract Automated vehicles are expected to navigate complex urban environments safely along with several non-cooperating agents. Therefore, accurate trajectory prediction is crucial for safe navigation and receives much interest. However, most existing studies mainly focus on models' prediction accuracy rather than their explainability. In this paper, we propose a Multimodal Trajectory Prediction Transformer (MTPT) model to retrieve the influencing factors of prediction and help understand the intrinsic mechanism of prediction. Especially in the MTPT model, we use a modified Swin Transformer with multiple prediction heads to carry out multimodal trajectory prediction. Numerical experiments confirm the MTPT model can capture the most critical input factors with the help of the attention method and thus improve prediction accuracy. The proposed model obtains state-of-the-art results even with limited training data. Moreover, the identified input factors are also in agreement with the human driving experience. This agreement indicates that the proposed model appropriately learns how to predict.
Explainable multimodal trajectory prediction using attention models
2022-07-28
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Online Contents | 2022
|Attention Based Vehicle Trajectory Prediction
IEEE | 2021
|TRAJECTORY PREDICTION USING EFFICIENT ATTENTION NEURAL NETWORKS
Europäisches Patentamt | 2023
|