Hierarchical Tactile Sensation Integration from Prosthetic Fingertips Enables Multi-Texture Surface Recognition.

Moaed A. Abd,Ou Bai,Maohua Lin,Rudy Paul,Leonel Lagos,Aparna Aravelli,Erik D. Engeberg

doi:10.3390/s21134324

Abstract

Multifunctional flexible tactile sensors could be useful to improve the control of prosthetic hands. To that end, highly stretchable liquid metal tactile sensors (LMS) were designed, manufactured via photolithography, and incorporated into the fingertips of a prosthetic hand. Three novel contributions were made with the LMS. First, individual fingertips were used to distinguish between different speeds of sliding contact with different surfaces. Second, differences in surface textures were reliably detected during sliding contact. Third, the capacity for hierarchical tactile sensor integration was demonstrated by using four LMS signals simultaneously to distinguish between ten complex multi-textured surfaces. Four different machine learning algorithms were compared for their successful classification capabilities: K-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and neural network (NN). The time-frequency features of the LMSs were extracted to train and test the machine learning algorithms. The NN generally performed the best at the speed and texture detection with a single finger and had a 99.2 ± 0.8% accuracy to distinguish between ten different multi-textured surfaces using four LMSs from four fingers simultaneously. The capability for hierarchical multi-finger tactile sensation integration could be useful to provide a higher level of intelligence for artificial hands.

Highlights

The sensation of touch for prosthetic hands is necessary to improve the upper limb amputee experience in everyday activities [1]
Sample data from the liquid metal tactile sensors (LMS) on the little finger sliding across texture texture 3 (T3) with three different speeds showed different characteristics in the time domain (Figure 7e–g)
Illustrative data from the LMS on the middle finger sliding across the four different textures (T1–texture 4 (T4)) with a speed of 20 mm/s showed noticeable differences in the sensor responses (Figure 7k–n)

Summary

Introduction

The sensation of touch for prosthetic hands is necessary to improve the upper limb amputee experience in everyday activities [1]. Available prosthetic hands like the i-limb Ultra (Figure 1), which has six degrees of freedom (DOF), and the BeBionic prosthetic hand that has five powered DOFs, demonstrate the trend of increasing prosthesis dexterity [2], yet these state of the art prosthetic limbs lack tactile sensation capabilities when interacting with the environment and manipulating objects. The absence of sensory feedback can lead to a frustrating problem when grasped objects are crushed or dropped since the amputee is not directly aware of the prosthetic fingertip forces after the afferent neural pathway is severed [3,4]. Human hand control strategies depend heavily on touch sensations for object manipulation [5]; people with upper limb amputation are missing tactile sensations. Significant research has been done on tactile sensors for artificial hands [6], but there is still a need for advances in lightweight, low-cost, robust multimodal tactile sensors [7].

Objectives

Methods

Results

Conclusion