SSL-Interactions: Pretext Tasks for Interactive Trajectory Prediction

TL;DR

This paper introduces SSL-Interactions, a set of pretext tasks designed to improve modeling of agent interactions in trajectory prediction, leading to better performance in multi-agent autonomous vehicle scenarios.

Contribution

It proposes novel interaction-aware pretext tasks and a data curation method to enhance interaction modeling in trajectory forecasting.

Findings

SSL-Interactions outperforms state-of-the-art methods by up to 8% in quantitative metrics.

The curated interaction-heavy dataset improves learning signals for interaction modeling.

New metrics effectively evaluate predictions in interactive scenes.

Abstract

This paper addresses motion forecasting in multi-agent environments, pivotal for ensuring safety of autonomous vehicles. Traditional as well as recent data-driven marginal trajectory prediction methods struggle to properly learn non-linear agent-to-agent interactions. We present SSL-Interactions that proposes pretext tasks to enhance interaction modeling for trajectory prediction. We introduce four interaction-aware pretext tasks to encapsulate various aspects of agent interactions: range gap prediction, closest distance prediction, direction of movement prediction, and type of interaction prediction. We further propose an approach to curate interaction-heavy scenarios from datasets. This curated data has two advantages: it provides a stronger learning signal to the interaction model, and facilitates generation of pseudo-labels for interaction-centric pretext tasks. We also propose three new metrics specifically designed to evaluate predictions in interactive scenes. Our empirical evaluations indicate SSL-Interactions outperforms state-of-the-art motion forecasting methods quantitatively with up to 8% improvement, and qualitatively, for interaction-heavy scenarios.