GeoDEx: A Unified Geometric Framework for Tactile Dexterous and Extrinsic Manipulation under Force Uncertainty

2025-05-01

AI Summary

Tactile robots face challenges in force-controlled manipulation due to sensor calibration difficulties and noise-induced inaccuracies in force measurement, hindering stable dexterous grasping and extrinsic manipulation. Method: This paper proposes a unified estimation, planning, and control framework grounded in geometric primitives—namely, planes, cones, and ellipsoids. It introduces these primitives systematically into tactile force uncertainty modeling for the first time, enabling a unified representation of both intrinsic (self-manipulation) and extrinsic (environment-interaction) tasks while reducing reliance on high-precision force sensors. The approach integrates geometric modeling, uncertainty-aware force estimation, second-order cone programming (SOCP)-based surrogate optimization, and closed-loop tactile feedback control. Results: Experiments demonstrate stable grasping and extrinsic manipulation across diverse objects. Compared to direct SOCP optimization, the method achieves a 14× speedup, significantly enhancing robustness and real-time performance.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Calibration & Uncertainty Quantification
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

Sense of touch that allows robots to detect contact and measure interaction forces enables them to perform challenging tasks such as grasping fragile objects or using tools. Tactile sensors in theory can equip the robots with such capabilities. However, accuracy of the measured forces is not on a par with those of the force sensors due to the potential calibration challenges and noise. This has limited the values these sensors can offer in manipulation applications that require force control. In this paper, we introduce GeoDEx, a unified estimation, planning, and control framework using geometric primitives such as plane, cone and ellipsoid, which enables dexterous as well as extrinsic manipulation in the presence of uncertain force readings. Through various experimental results, we show that while relying on direct inaccurate and noisy force readings from tactile sensors results in unstable or failed manipulation, our method enables successful grasping and extrinsic manipulation of different objects. Additionally, compared to directly running optimization using SOCP (Second Order Cone Programming), planning and force estimation using our framework achieves a 14x speed-up.

Problem

• Enables dexterous manipulation under force uncertainty
• Improves tactile sensor reliability for force control
• Achieves faster planning than SOCP optimization

Innovation

• Unified geometric framework for tactile manipulation
• Uses geometric primitives for force estimation
• Achieves 14x speed-up over SOCP optimization