Large-area Tomographic Tactile Skin with Air Pressure Sensing for Improved Force Estimation

2025-03-17

AI Summary

This work addresses the challenge of simultaneously localizing and quantifying multi-point contact forces on large-area tactile skins without external calibration. We propose a novel tomographic tactile perception paradigm that synergistically integrates electrical impedance tomography (EIT) with pneumatic pressure sensing. Our approach establishes a deep coupling reconstruction framework that jointly maps EIT-derived spatial conductivity distributions to total pneumatic force, leveraging deep learning–based image reconstruction, contact region segmentation, and conductivity-driven force distribution. Crucially, this enables joint inversion of contact locations and individual force magnitudes without complex external calibration. Experimental results demonstrate mean force estimation errors of 15.1% for single-point and 20.1% for multi-point contacts—substantially reducing reliance on large-scale calibration datasets. The method offers a scalable, low-calibration-overhead solution for practical deployment of flexible tactile skins.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Biometrics, Face, Gesture & Pose

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Security and Privacy: Large-scale security measurements
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

This paper presents a dual-channel tactile skin that integrates Electrical Impedance Tomography (EIT) with air pressure sensing to achieve accurate multi-contact force detection. The EIT layer provides spatial contact information, while the air pressure sensor delivers precise total force measurement. Our framework combines these complementary modalities through: deep learning-based EIT image reconstruction, contact area segmentation, and force allocation based on relative conductivity intensities from EIT. The experiments demonstrated 15.1% average force estimation error in single-contact scenarios and 20.1% in multi-contact scenarios without extensive calibration data requirements. This approach effectively addresses the challenge of simultaneously localizing and quantifying multiple contact forces without requiring complex external calibration setups, paving the way for practical and scalable soft robotic skin applications.

Problem

• Accurate multi-contact force detection using dual-channel tactile skin.
• Combining EIT and air pressure sensing for force measurement.
• Reducing calibration needs for force localization and quantification.

Innovation

• Dual-channel tactile skin with EIT and air pressure sensing
• Deep learning-based EIT image reconstruction and force allocation
• Accurate multi-contact force detection without complex calibration