Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Robotics
This work addresses the problem of enhancing robot control through better embodied reasoning for robotics applications, representing an incremental improvement over existing methods.
The paper tackles the problem of improving embodied reasoning for robot control by addressing limitations of supervised fine-tuning, such as heuristic datasets and reduced generalization, and introduces Robot-R1, a reinforcement learning framework that outperforms SFT methods and even GPT-4o on tasks like spatial and primitive movement reasoning with a 7B parameter model.
Large Vision-Language Models (LVLMs) have recently shown great promise in advancing robotics by combining embodied reasoning with robot control. A common approach involves training on embodied reasoning tasks related to robot control using Supervised Fine-Tuning (SFT). However, SFT datasets are often heuristically constructed and not explicitly optimized for improving robot control. Furthermore, SFT often leads to issues such as catastrophic forgetting and reduced generalization performance. To address these limitations, we introduce Robot-R1, a novel framework that leverages reinforcement learning to enhance embodied reasoning specifically for robot control. Robot-R1 learns to predict the next keypoint state required for task completion, conditioned on the current scene image and environment metadata derived from expert demonstrations. Inspired by the DeepSeek-R1 learning approach, Robot-R1 samples reasoning-based responses and reinforces those that lead to more accurate predictions. Our experiments show that models trained with Robot-R1 outperform SFT methods on embodied reasoning tasks. Despite having only 7B parameters, Robot-R1 even surpasses GPT-4o on reasoning tasks related to low-level action control, such as spatial and primitive movement reasoning.