Trajectory Prediction using Equivariant Continuous Convolution

TL;DR

This paper introduces ECCO, a novel trajectory prediction model using rotationally-equivariant continuous convolutions, which enhances physical consistency, accuracy, and sample efficiency by leveraging symmetry insights from fluid dynamics.

Contribution

The paper presents ECCO, a new model that incorporates rotational equivariance for trajectory prediction, improving physical realism and efficiency over existing methods.

Findings

ECCO achieves competitive accuracy with fewer parameters.

ECCO generalizes well from limited data and various orientations.

ECCO produces more physically consistent trajectory predictions.

Abstract

Trajectory prediction is a critical part of many AI applications, for example, the safe operation of autonomous vehicles. However, current methods are prone to making inconsistent and physically unrealistic predictions. We leverage insights from fluid dynamics to overcome this limitation by considering internal symmetry in real-world trajectories. We propose a novel model, Equivariant Continous COnvolution (ECCO) for improved trajectory prediction. ECCO uses rotationally-equivariant continuous convolutions to embed the symmetries of the system. On both vehicle and pedestrian trajectory datasets, ECCO attains competitive accuracy with significantly fewer parameters. It is also more sample efficient, generalizing automatically from few data points in any orientation. Lastly, ECCO improves generalization with equivariance, resulting in more physically consistent predictions. Our method provides a fresh perspective towards increasing trust and transparency in deep learning models.