CoIRL-AD: Collaborative-Competitive Imitation-Reinforcement Learning in Latent World Models for Autonomous Driving
This work addresses generalization and safety issues in autonomous driving, representing an incremental improvement over existing hybrid methods.
The paper tackles the poor generalization of end-to-end autonomous driving models by proposing CoIRL-AD, a competitive dual-policy framework that combines imitation and reinforcement learning, resulting in an 18% reduction in collision rate on the nuScenes dataset.
End-to-end autonomous driving models trained solely with imitation learning (IL) often suffer from poor generalization. In contrast, reinforcement learning (RL) promotes exploration through reward maximization but faces challenges such as sample inefficiency and unstable convergence. A natural solution is to combine IL and RL. Moving beyond the conventional two-stage paradigm (IL pretraining followed by RL fine-tuning), we propose CoIRL-AD, a competitive dual-policy framework that enables IL and RL agents to interact during training. CoIRL-AD introduces a competition-based mechanism that facilitates knowledge exchange while preventing gradient conflicts. Experiments on the nuScenes dataset show an 18% reduction in collision rate compared to baselines, along with stronger generalization and improved performance on long-tail scenarios. Code is available at: https://github.com/SEU-zxj/CoIRL-AD.