ArrayBot: Reinforcement Learning for Generalizable Distributed Manipulation through Touch

We present ArrayBot, a distributed manipulation system consisting of a 16 × 16 array of vertically sliding pillars integrated with tactile sensors. Functionally, ArrayBot is designed to simultaneously support, perceive, and manipulate the tabletop objects. Towards generalizable distributed manipulation, we leverage reinforcement learning (RL) algorithms for the automatic discovery of control policies. In the face of the massively redundant actions, we propose to reshape the action space by considering the spatially local action patch and the low-frequency actions in the frequency domain. With this reshaped action space, we train RL agents that can relocate diverse objects through tactile observations only. Intriguingly, we find that the discovered policy can not only generalize to unseen object shapes in the simulator but also have the ability to transfer to the physical robot without any sim-to-real fine-tuning. |Leveraging the deployed policy, we derive more real-world manipulation skills on ArrayBot to further illustrate the distinctive merits of our proposed system.

Authors

• Zhengrong Xue
• Han Zhang 0058
• Jingwen Cheng
• Zhengmao He
• Yuanchen Ju
• Changyi Lin
• Gu Zhang
• Huazhe Xu

Keywords

• Shape
• Frequency-domain analysis
• Tactile sensors
• Reinforcement learning
• Aerospace electronics
• Sensors
• Faces
• Frequency Domain
• Object Shape
• Tactile Sensor
• Reinforcement Learning Algorithm
• Reinforcement Learning Agent
• Physical Robot
• Unseen Objects
• Actuator
• Spatial Domain
• Robotic Arm
• Object Position
• Central Objective
• Atomic Units
• Physical Simulation
• Discrete Cosine Transform
• Inductive Bias
• Model-free Reinforcement Learning
• Proximal Policy Optimization
• Modular Units
• Frequency Transformer
• Force Sensitive Resistors
• Motion Primitives