Watch Less, Feel More: Sim-to-Real RL for Generalizable Articulated Object Manipulation via Motion Adaptation and Impedance Control

2025-02-20

AI Summary

To address the challenge of zero-shot sim-to-real transfer for articulated object manipulation, this paper proposes an end-to-end reinforcement learning control framework. The method integrates motion-adaptive control and variable-impedance modulation to enable smooth, dexterous manipulation; introduces a history-observation-driven joint inference mechanism for motion and object properties; and designs a dual-perception reward function that jointly considers task completion and motion quality. It employs low-dimensional state representations (excluding end-to-end RGB-D or point-cloud inputs), sequence-based observation encoding, and highly randomized simulation training. Evaluated on real robotic arms, the approach achieves an 84% zero-shot success rate across unseen articulated objects—including doors, drawers, and faucets—without any real-world fine-tuning. This significantly outperforms prior methods and, for the first time, empirically validates the feasibility of high-generalization, cross-object zero-shot manipulation in real-world settings.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Articulated object manipulation poses a unique challenge compared to rigid object manipulation as the object itself represents a dynamic environment. In this work, we present a novel RL-based pipeline equipped with variable impedance control and motion adaptation leveraging observation history for generalizable articulated object manipulation, focusing on smooth and dexterous motion during zero-shot sim-to-real transfer. To mitigate the sim-to-real gap, our pipeline diminishes reliance on vision by not leveraging the vision data feature (RGBD/pointcloud) directly as policy input but rather extracting useful low-dimensional data first via off-the-shelf modules. Additionally, we experience less sim-to-real gap by inferring object motion and its intrinsic properties via observation history as well as utilizing impedance control both in the simulation and in the real world. Furthermore, we develop a well-designed training setting with great randomization and a specialized reward system (task-aware and motion-aware) that enables multi-staged, end-to-end manipulation without heuristic motion planning. To the best of our knowledge, our policy is the first to report 84% success rate in the real world via extensive experiments with various unseen objects.

Problem

• Develops RL for articulated object manipulation.
• Reduces sim-to-real gap via motion adaptation.
• Achieves 84% success rate in real-world tests.

Innovation

• Sim-to-Real RL pipeline
• Variable impedance control
• Observation history utilization