Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Robotics

TL;DR

Robot-R1 introduces reinforcement learning to improve embodied reasoning in robotics, addressing limitations of supervised fine-tuning by enhancing control and generalization, and outperforming existing methods including GPT-4o on key tasks.

Contribution

The paper presents Robot-R1, a reinforcement learning framework that improves embodied reasoning for robot control, overcoming SFT limitations and introducing a new benchmark for evaluation.

Findings

Robot-R1 outperforms SFT methods on embodied reasoning tasks.

Robot-R1 surpasses GPT-4o on low-level action control reasoning.

Models trained with Robot-R1 demonstrate enhanced generalization and accuracy.

Abstract

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. To rigorously evaluate Robot-R1, we also introduce a new benchmark that demands the diverse embodied reasoning capabilities for the task. 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 movement reasoning.