Generalizable Trajectory Prediction via Inverse Reinforcement Learning with Mamba-Graph Architecture
This work addresses trajectory prediction for autonomous driving systems, offering improved cross-scenario adaptability.
The paper tackles the challenge of accurate driving behavior modeling in complex traffic scenarios by proposing a novel Inverse Reinforcement Learning framework that infers diverse reward functions to capture human-like decision-making. The method achieves 2 times higher generalization performance to unseen scenarios compared to other IRL-based methods while outperforming various popular approaches in prediction accuracy.
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 the encoder-decoder architecture that combines 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 prediction accuracy but also achieves 2 times higher generalization performance to unseen scenarios compared to other IRL-based method.