EvolveGraph: Multi-Agent Trajectory Prediction with Dynamic Relational Reasoning

TL;DR

EvolveGraph is a flexible multi-agent trajectory prediction framework that models dynamic interactions with evolving relational graphs, providing multi-modal forecasts and achieving state-of-the-art accuracy on synthetic and real-world datasets.

Contribution

The paper introduces EvolveGraph, a novel framework that explicitly models evolving interaction graphs for multi-agent trajectory prediction, incorporating dynamic relational reasoning and a double-stage training pipeline.

Findings

Achieves state-of-the-art prediction accuracy on benchmark datasets.

Effectively models dynamic, evolving interactions among heterogeneous agents.

Provides multi-modal trajectory predictions considering future uncertainty.

Abstract

Multi-agent interacting systems are prevalent in the world, from pure physical systems to complicated social dynamic systems. In many applications, effective understanding of the situation and accurate trajectory prediction of interactive agents play a significant role in downstream tasks, such as decision making and planning. In this paper, we propose a generic trajectory forecasting framework (named EvolveGraph) with explicit relational structure recognition and prediction via latent interaction graphs among multiple heterogeneous, interactive agents. Considering the uncertainty of future behaviors, the model is designed to provide multi-modal prediction hypotheses. Since the underlying interactions may evolve even with abrupt changes, and different modalities of evolution may lead to different outcomes, we address the necessity of dynamic relational reasoning and adaptively evolving the interaction graphs. We also introduce a double-stage training pipeline which not only improves training efficiency and accelerates convergence, but also enhances model performance. The proposed framework is evaluated on both synthetic physics simulations and multiple real-world benchmark datasets in various areas. The experimental results illustrate that our approach achieves state-of-the-art performance in terms of prediction accuracy.