Tracing Energy Flow: Learning Tactile-Based Grasping Force Control to Reduce Slippage in Dynamic Object Interaction
Cheng-Yu Kuo, Hirofumi Shin, Takamitsu Matsubara
AI summary
Problem
Regulating grasping force to prevent slippage during dynamic object interaction is difficult due to unknown object properties, complex multi-finger dynamics, and unreliable external sensing. Existing methods struggle to infer slip-aware stability without explicit supervision or visual cues.
Approach
The method models the object as a virtual energy container, using tactile sensing to compare applied power with retained energy to detect slippage. This energy inconsistency signal drives a model-based reinforcement learning framework that optimizes grasping force in real time via probabilistic model predictive control.
Key results
- Physics-informed energy abstraction infers slippage from tactile energy inconsistency
- Integration with model-based reinforcement learning enables real-time grasping force optimization
- Successfully learns grasp control from scratch within minutes in simulation and hardware
- Effectively reduces slippage across diverse objects and motions without external sensing or prior object knowledge
Why it matters
Enables robust, tactile-only robotic manipulation in real-world scenarios where vision fails and object properties are unknown, advancing autonomous dexterous manipulation.
Abstract
Regulating grasping force to reduce slippage during dynamic object interaction remains a fundamental challenge in robotic manipulation, especially when objects are manipulated by multiple rolling contacts, have unknown properties (such as mass or surface conditions), and when external sensing is unreliable. In contrast, humans can quickly regulate grasping force by touch, even without visual cues. Inspired by this ability, we aim to enable robotic hands to rapidly explore objects and learn tactile-driven grasping force control under motion and limited sensing. We propose a physics-informed energy abstraction that models the object as a virtual energy container. The inconsistency between the fingers’ applied power and the object’s retained energy provides a physically grounded signal for inferring slip-aware stability. Building on this abstraction, we employ model-based learning and planning to efficiently model energy dynamics from tactile sensing and perform real-time grasping force optimization. Experiments in both simulation and hardware demonstrate that our method can learn grasping force control from scratch within minutes, effectively reduce slippage, and extend grasp duration across diverse motion-object pairs, all without relying on external sensing or prior object knowledge.