Improving Trajectory Prediction in Dynamic Multi-Agent Environment by Dropping Waypoints

TL;DR

This paper introduces Temporal Waypoint Dropping (TWD), a novel training technique that improves trajectory prediction in dynamic multi-agent environments by handling missing waypoint data through stochastic dropping during training.

Contribution

The paper proposes TWD, a new method that enhances trajectory prediction accuracy by explicitly modeling temporal dependencies and robustness to missing data during training.

Findings

Significant improvement in prediction accuracy on three datasets.

Enhanced robustness to missing waypoint data.

Complementary to existing trajectory prediction methods.

Abstract

The inherently diverse and uncertain nature of trajectories presents a formidable challenge in accurately modeling them. Motion prediction systems must effectively learn spatial and temporal information from the past to forecast the future trajectories of the agent. Many existing methods learn temporal motion via separate components within stacked models to capture temporal features. Furthermore, prediction methods often operate under the assumption that observed trajectory waypoint sequences are complete, disregarding scenarios where missing values may occur, which can influence their performance. Moreover, these models may be biased toward particular waypoint sequences when making predictions. We propose a novel approach called Temporal Waypoint Dropping (TWD) that explicitly incorporates temporal dependencies during the training of a trajectory prediction model. By stochastically dropping waypoints from past observed trajectories, the model is forced to learn the underlying temporal representation from the remaining waypoints, resulting in an improved model. Incorporating stochastic temporal waypoint dropping into the model learning process significantly enhances its performance in scenarios with missing values. Experimental results demonstrate our approach's substantial improvement in trajectory prediction capabilities. Our approach can complement existing trajectory prediction methods to improve their prediction accuracy. We evaluate our proposed approach on three datasets: NBA Sports VU, ETH-UCY, and TrajNet++.