MAP-Former: Multi-Agent-Pair Gaussian Joint Prediction

TL;DR

This paper introduces MAP-Former, a novel model that predicts agent-pair covariance matrices to generate joint Gaussian PDFs, enhancing traffic risk assessment by capturing inter-agent dependencies.

Contribution

It presents a new approach to predict agent-pair covariances for joint trajectory modeling, addressing limitations of existing marginal prediction methods.

Findings

Successfully predicts agent-pair covariance matrices.

Enables modeling of joint PDFs for all agent pairs in a scene.

Provides a foundation for improved risk assessment in traffic scenarios.

Abstract

There is a gap in risk assessment of trajectories between the trajectory information coming from a traffic motion prediction module and what is actually needed. Closing this gap necessitates advancements in prediction beyond current practices. Existing prediction models yield joint predictions of agents' future trajectories with uncertainty weights or marginal Gaussian probability density functions (PDFs) for single agents. Although, these methods achieve high accurate trajectory predictions, they only provide little or no information about the dependencies of interacting agents. Since traffic is a process of highly interdependent agents, whose actions directly influence their mutual behavior, the existing methods are not sufficient to reliably assess the risk of future trajectories. This paper addresses that gap by introducing a novel approach to motion prediction, focusing on predicting agent-pair covariance matrices in a ``scene-centric'' manner, which can then be used to model Gaussian joint PDFs for all agent-pairs in a scene. We propose a model capable of predicting those agent-pair covariance matrices, leveraging an enhanced awareness of interactions. Utilizing the prediction results of our model, this work forms the foundation for comprehensive risk assessment with statistically based methods for analyzing agents' relations by their joint PDFs.