Lump localisation through a deformation-based tactile feedback system using a biologically inspired finger sensor

Pipe, Tony; Roke, Calum; Melhuish, Chris; Pipe, Anthony G.; Drury, David; Chorley, Craig

Abstract

The ability to localise harder areas in soft tissues is often desired during robot-assisted surgical operations. A deformation-based tactile feedback system was tested for the detection of objects within soft tissues, after being chosen over common pressure-based designs. This system uses a biologically inspired sensor that offers a new finger-like approach to tactile sensing. A tactile shape display developed from previous successful designs was used to output the sensed tactile information. Using the tactile feedback system on a mechanical teleoperated device, test subjects palpated a number of artificial tissue models to locate objects of varying stiffness. The addition of the tactile feedback system improved the detection of the objects from 64% to 98%, reduced the localisation error from 18 to 11 mm, and also decreased the time the users spent palpating the tissue from 55 to 37 s. This demonstrates that a deformation-based tactile feedback system can be used to successfully locate hard embedded objects within soft tissue, with a significant improvement over force and visual feedback alone. During testing, it was found that the users were able to more accurately locate the softest embedded objects compared to stiffer ones. Reasons for this observation are discussed. © 2012 Elsevier B.V. All rights reserved.

• Journal Article Type: Article
• Publication Date: Nov 1, 2012
• Deposit Date: Feb 28, 2013
• Journal: Robotics and Autonomous Systems
• Print ISSN: 0921-8890
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 60
• Issue: 11
• Pages: 1442-1448
• DOI: https://doi.org/10.1016/j.robot.2012.05.002
• Keywords: biologically inspired finger sensor, lump localisation
• Public URL: https://uwe-repository.worktribe.com/output/942542
• Publisher URL: http://dx.doi.org/10.1016/j.robot.2012.05.002
• Contract Date: Mar 20, 2016