TacFinRay: Soft Tactile Fin-Ray Finger with Indirect Tactile Sensing for Robust Grasping

2025-12-06

AI Summary

To address the limited direct tactile sensing in soft robotic fingers caused by sensor placement away from the contact interface, this work proposes a tactile-enabled Fin-Ray finger based on indirect perception. Contact-induced deformation is mechanically transmitted via a hinge mechanism to a distal TacTip vision sensor; an integrated camera and lightweight convolutional neural network jointly estimate contact location and indentation depth. The key innovation lies in the physical decoupling of the sensing module from the soft actuation structure, enabling both high accuracy (0.1 mm depth error, 2 mm position error) and strong generalization across arbitrarily shaped objects for robust grasping. Experiments demonstrate significantly improved placement accuracy in pick-and-place tasks under uncertain initial object poses, validating the design’s practicality and scalability in complex, unstructured environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Biometrics, Face, Gesture & Pose
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

We present a tactile-sensorized Fin-Ray finger that enables simultaneous detection of contact location and indentation depth through an indirect sensing approach. A hinge mechanism is integrated between the soft Fin-Ray structure and a rigid sensing module, allowing deformation and translation information to be transferred to a bottom crossbeam upon which are an array of marker-tipped pins based on the biomimetic structure of the TacTip vision-based tactile sensor. Deformation patterns captured by an internal camera are processed using a convolutional neural network to infer contact conditions without directly sensing the finger surface. The finger design was optimized by varying pin configurations and hinge orientations, achieving 0.1,mm depth and 2mm location-sensing accuracies. The perception demonstrated robust generalization to various indenter shapes and sizes, which was applied to a pick-and-place task under uncertain picking positions, where the tactile feedback significantly improved placement accuracy. Overall, this work provides a lightweight, flexible, and scalable tactile sensing solution suitable for soft robotic structures where the sensing needs situating away from the contact interface.

Problem

• Develops a tactile Fin-Ray finger for contact detection
• Uses indirect sensing to infer contact location and depth
• Enhances robotic grasping accuracy with tactile feedback

Innovation

• Indirect tactile sensing via hinge mechanism
• Marker-tipped pins array for deformation capture
• CNN processing for contact inference without direct sensing