IPCC-TP: Utilizing Incremental Pearson Correlation Coefficient for Joint Multi-Agent Trajectory Prediction

TL;DR

This paper introduces IPCC-TP, a relevance-aware module based on Incremental Pearson Correlation Coefficient, which enhances multi-agent trajectory prediction by better modeling social interactions, leading to significant performance improvements.

Contribution

The paper proposes a novel IPCC-TP module that learns pairwise joint Gaussian distributions for multi-agent interactions, improving upon previous graph-based and attention mechanisms.

Findings

IPCC-TP significantly outperforms baseline methods on nuScenes dataset.

The module effectively captures complex social interactions in multi-agent scenarios.

Extensive experiments validate the robustness and accuracy of the proposed approach.

Abstract

Reliable multi-agent trajectory prediction is crucial for the safe planning and control of autonomous systems. Compared with single-agent cases, the major challenge in simultaneously processing multiple agents lies in modeling complex social interactions caused by various driving intentions and road conditions. Previous methods typically leverage graph-based message propagation or attention mechanism to encapsulate such interactions in the format of marginal probabilistic distributions. However, it is inherently sub-optimal. In this paper, we propose IPCC-TP, a novel relevance-aware module based on Incremental Pearson Correlation Coefficient to improve multi-agent interaction modeling. IPCC-TP learns pairwise joint Gaussian Distributions through the tightly-coupled estimation of the means and covariances according to interactive incremental movements. Our module can be conveniently embedded into existing multi-agent prediction methods to extend original motion distribution decoders. Extensive experiments on nuScenes and Argoverse 2 datasets demonstrate that IPCC-TP improves the performance of baselines by a large margin.