Generalizable Trajectory Prediction via Inverse Reinforcement Learning with Mamba-Graph Architecture

TL;DR

This paper introduces a novel IRL framework with Mamba-Graph architecture for trajectory prediction, enhancing cross-scenario adaptability and out-of-distribution generalization in complex traffic environments.

Contribution

It proposes a new IRL-based approach using Mamba blocks and graph attention networks to improve trajectory prediction and generalization across diverse traffic scenarios.

Findings

Outperforms existing methods in prediction accuracy.

Achieves 2.3 times higher generalization to unseen scenarios.

Demonstrates robustness in complex urban traffic environments.

Abstract

Accurate driving behavior modeling is fundamental to safe and efficient trajectory prediction, yet remains challenging in complex traffic scenarios. This paper presents a novel Inverse Reinforcement Learning (IRL) framework that captures human-like decision-making by inferring diverse reward functions, enabling robust cross-scenario adaptability. The learned reward function is utilized to maximize the likelihood of output by integrating Mamba blocks for efficient long-sequence dependency modeling with graph attention networks to encode spatial interactions among traffic agents. Comprehensive evaluations on urban intersections and roundabouts demonstrate that the proposed method not only outperforms various popular approaches in terms of prediction accuracy but also achieves 2.3 times higher generalization performance to unseen scenarios compared to other baselines, achieving adaptability in Out-of-Distribution settings that is competitive with fine-tuning.