RedMotion: Motion Prediction via Redundancy Reduction

TL;DR — Transformer model for motion prediction in self-driving that incorporates two types of redundancy reduction

Abstract

Predicting the future motion of traffic agents is vital for self-driving vehicles to ensure their safe operation. We introduce RedMotion, a transformer model for motion prediction that incorporates two types of redundancy reduction. The first type of redundancy reduction is induced by an internal transformer decoder and reduces a variable-sized set of road environment tokens, such as road graphs with agent data, to a fixed-sized embedding. The second type of redundancy reduction is a self-supervised learning objective and applies the redundancy reduction principle to embeddings generated from augmented views of road environments. Our experiments reveal that our representation learning approach can outperform PreTraM, Traj-MAE, and GraphDINO in a semi-supervised setting. Our RedMotion model achieves results that are competitive with those of Scene Transformer or MTR++. We provide an anonymized open source implementation that is accessible via Colab: https://colab.research.google.com/drive/16pwsmOTYdPpbNWf2nm1olXcx1ZmsXHB8