Part-Guided 3D RL for Sim2Real Articulated Object Manipulation

• Published in: IEEE robotics and automation letters Vol. 8; no. 11; pp. 7178 - 7185
• Main Authors: Xie, Pengwei, Chen, Rui, Chen, Siang, Qin, Yuzhe, Xiang, Fanbo, Sun, Tianyu, Xu, Jing, Wang, Guijin, Su, Hao
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.11.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Deep learning in grasping and manipulation; Frame design; Image segmentation; Learning; Point cloud compression; reinforcement learning; RGB-D perception; Robots; Task analysis; Three-dimensional displays; Uncertainty; Visualization
• ISSN: 2377-3766; 2377-3766
• DOI: 10.1109/LRA.2023.3313063

Manipulating unseen articulated objects through visual feedback is a critical but challenging task for real robots. Existing learning-based solutions mainly focus on visual affordance learning or other pre-trained visual models to guide manipulation policies, which face challenges for novel instances in real-world scenarios. In this letter, we propose a novel part-guided 3D RL framework, which can learn to manipulate articulated objects without demonstrations. We combine the strengths of 2D segmentation and 3D RL to improve the efficiency of RL policy training. To improve the stability of the policy on real robots, we design a Frame-consistent Uncertainty-aware Sampling (FUS) strategy to get a condensed and hierarchical 3D representation. In addition, a single versatile RL policy can be trained on multiple articulated object manipulation tasks simultaneously in simulation and shows great generalizability to novel categories and instances. Experimental results demonstrate the effectiveness of our framework in both simulation and real-world settings.