Tracing Energy Flow: Learning Tactile-based Grasping Force Control to Prevent Slippage in Dynamic Object Interaction

TL;DR

This paper introduces a tactile-driven, physics-informed approach for robotic grasping that learns to control grasping force in real-time, reducing slippage during dynamic interactions without external sensing or prior object knowledge.

Contribution

It proposes a novel energy abstraction model and model-based learning framework enabling robots to learn grasping force control quickly from tactile feedback alone.

Findings

Effective slip reduction in simulation and hardware

Rapid learning within minutes from scratch

Extended grasp duration across diverse objects

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.