CHARMS: A Cognitive Hierarchical Agent for Reasoning and Motion Stylization in Autonomous Driving
This work addresses the challenge of improving interaction fidelity and diversity in autonomous driving systems, representing an incremental advancement by building on existing game theory and training methods.
The paper tackles the problem of insufficient interactivity and behavioral diversity in autonomous driving decision-making by proposing CHARMS, a cognitive hierarchical agent that uses Level-k game theory and a two-stage training pipeline to enable human-like reasoning and diverse behaviors, with experimental results showing it can make intelligent driving decisions and generate realistic scenarios.
To address the challenge of insufficient interactivity and behavioral diversity in autonomous driving decision-making, this paper proposes a Cognitive Hierarchical Agent for Reasoning and Motion Stylization (CHARMS). By leveraging Level-k game theory, CHARMS captures human-like reasoning patterns through a two-stage training pipeline comprising reinforcement learning pretraining and supervised fine-tuning. This enables the resulting models to exhibit diverse and human-like behaviors, enhancing their decision-making capacity and interaction fidelity in complex traffic environments. Building upon this capability, we further develop a scenario generation framework that utilizes the Poisson cognitive hierarchy theory to control the distribution of vehicles with different driving styles through Poisson and binomial sampling. Experimental results demonstrate that CHARMS is capable of both making intelligent driving decisions as an ego vehicle and generating diverse, realistic driving scenarios as environment vehicles. The code for CHARMS is released at https://github.com/chuduanfeng/CHARMS.