GoIRL: Graph-Oriented Inverse Reinforcement Learning for Multimodal Trajectory Prediction

TL;DR

This paper introduces GoIRL, a graph-oriented inverse reinforcement learning framework for multimodal trajectory prediction in autonomous driving, achieving state-of-the-art results and better generalization than supervised methods.

Contribution

We propose a novel IRL-based predictor with vectorized context representations and a hierarchical trajectory generator for improved multimodal prediction accuracy.

Findings

State-of-the-art performance on Argoverse and nuScenes benchmarks

Superior generalization compared to supervised models

Effective integration of lane-graph features into IRL framework

Abstract

Trajectory prediction for surrounding agents is a challenging task in autonomous driving due to its inherent uncertainty and underlying multimodality. Unlike prevailing data-driven methods that primarily rely on supervised learning, in this paper, we introduce a novel Graph-oriented Inverse Reinforcement Learning (GoIRL) framework, which is an IRL-based predictor equipped with vectorized context representations. We develop a feature adaptor to effectively aggregate lane-graph features into grid space, enabling seamless integration with the maximum entropy IRL paradigm to infer the reward distribution and obtain the policy that can be sampled to induce multiple plausible plans. Furthermore, conditioned on the sampled plans, we implement a hierarchical parameterized trajectory generator with a refinement module to enhance prediction accuracy and a probability fusion strategy to boost prediction confidence. Extensive experimental results showcase our approach not only achieves state-of-the-art performance on the large-scale Argoverse & nuScenes motion forecasting benchmarks but also exhibits superior generalization abilities compared to existing supervised models.