GelBelt: A Vision-based Tactile Sensor for Continuous Sensing of Large Surfaces

2025-01-09

IEEE Robotics and Automation Letters

AI Summary

Existing tactile sensors suffer from limited field-of-view, susceptibility to wear, and unstable sliding motion, hindering large-area surface inspection (e.g., post-repair quality assessment). To address this, we propose an elastic conveyor-belt-style vision-based tactile sensor. Our method integrates dual-wheel actuation with real-time visual feedback to enable non-destructive, continuous, large-scale sliding perception. By combining structured-light projection, high-speed imaging, surface normal estimation, and multi-frame geometric fusion, it overcomes the limitations of static or small-region tactile sensing. Experiments demonstrate that at a scanning speed of 45 mm/s, the mean dot product of estimated and ground-truth surface normals exceeds 0.92, significantly improving accuracy, real-time performance, and robustness in large-area surface topography reconstruction. The core contributions are: (1) the first extensible conveyor-belt mechanical architecture for tactile sensing, and (2) a dynamically tight-coupled vision–tactile perception paradigm.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Scanning large-scale surfaces is widely demanded in surface reconstruction applications and detecting defects in industries' quality control and maintenance stages. Traditional vision-based tactile sensors have shown promising performance in high-resolution shape reconstruction while suffering limitations such as small sensing areas or susceptibility to damage when slid across surfaces, making them unsuitable for continuous sensing on large surfaces. To address these shortcomings, we introduce a novel vision-based tactile sensor designed for continuous surface sensing applications. Our design uses an elastomeric belt and two wheels to continuously scan the target surface. The proposed sensor showed promising results in both shape reconstruction and surface fusion, indicating its applicability. The dot product of the estimated and reference surface normal map is reported over the sensing area and for different scanning speeds. Results indicate that the proposed sensor can rapidly scan large-scale surfaces with high accuracy at speeds up to 45 mm/s.

Problem

• Tactile Sensors
• Limited Field of View
• Durability Issues

Innovation

• GelBelt
• Tactile Sensor
• Large-Area Monitoring