Built Different: Tactile Perception to Overcome Cross-Embodiment Capability Differences in Collaborative Manipulation

2024-09-23

arXiv.org

AI Summary

Commercial robots lacking torque feedback and non-impedance controllers struggle to adapt to human motion and perform force-rich collaborative manipulation. Method: This paper proposes a tactile-only cross-morphology collaborative strategy transfer framework. By fusing multi-point tactile signals, it enables implicit human force-intent estimation and policy distillation, transferring impedance-based collaborative control policies from torque-sensing impedance-controlled robots to purely position-controlled planar rail robots without joint torque sensing. Contribution/Results: To our knowledge, this is the first zero-shot cross-hardware behavioral transfer method that requires no joint torque sensors and no low-level controller modification. Experiments demonstrate that the target robot can estimate human motion intent in real time using tactile feedback alone, achieving stable compliance and high-precision manipulation in spatial cooperative transport tasks—effectively bridging capability gaps imposed by embodiment heterogeneity.

Technology Category

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

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: The sharing economy
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

Abstract

Tactile sensing is a powerful means of implicit communication between a human and a robot assistant. In this paper, we investigate how tactile sensing can transcend cross-embodiment differences across robotic systems in the context of collaborative manipulation. Consider tasks such as collaborative object carrying where the human-robot interaction is force rich. Learning and executing such skills requires the robot to comply to the human and to learn behaviors at the joint-torque level. However, most robots do not offer this compliance or provide access to their joint torques. To address this challenge, we present an approach that uses tactile sensors to transfer policies from robots with these capabilities to those without. We show how our method can enable a cooperative task where a robot and human must work together to maneuver objects through space. We first demonstrate the skill on an impedance control-capable robot equipped with tactile sensing, then show the positive transfer of the tactile policy to a planar prismatic robot that is only capable of position control and does not come equipped with any sort of force/torque feedback, yet is able to comply to the human motions only using tactile feedback. Further details and videos can be found on our project website at https://www.mmintlab.com/research/tactile-collaborative/.

Problem

• Using tactile sensing to bridge robot embodiment differences
• Transferring policies from torque-capable to torque-lacking robots
• Enabling collaborative manipulation across diverse robotic platforms

Innovation

• Tactile sensors enable cross-embodiment policy transfer
• Behavior cloning transfers policies to torque-limited robots
• Decomposed tactile shear-field representation improves success rates