Texture Recognition Using Force Sensitive Resistors
This paper presents the results of an experiment that investigates the presence of cues in the signal generated by a low-cost Force Sensitive Resistor (FSR) to recognise surface texture. The sensor is moved across the surface and the data is analysed to i
- PDF / 2,502,022 Bytes
- 7 Pages / 439.37 x 666.142 pts Page_size
- 42 Downloads / 179 Views
stract. This paper presents the results of an experiment that investigates the presence of cues in the signal generated by a low-cost Force Sensitive Resistor (FSR) to recognise surface texture. The sensor is moved across the surface and the data is analysed to investigate the presence of any patterns. We show that the signal contains enough information to recognise at least one sample surface.
Keywords: Texture recognition sensing
1
·
Force Sensitive Resistor
·
Active
Introduction
Humans perform a repetitive lateral rubbing motion across a surface to feel its texture, an action known as Lateral Motion Exploratory Procedure [5]. The physiology of the human sense of touch suggests that the information the human brain receives during this motion is coming from force sensory elements embedded in the skin and encoded by frequency modulation [4]. Researchers were able to interface a Force Sensitive Resistor (FSR) sensor, installed on a fingertip of a prosthetic hand, with the user’s nerves [8]. The user reported the ability to perceive “texture” of surfaces. This suggests that the single point force data acquired by the FSR hold enough information to perceive surface texture. We propose that the same ability can be replicated in an artificial system using the same sensor. It would be particularly useful to achieve this ability using FSRs due to their low cost and low thickness that enables superficial installation on robotic hands.
2
Related Work
In [3], researchers constructed a low-profile fabric tactile sensor which was able to differentiate between three surface textures. The sensor was moved across the surface with constant velocity and contact pressure. The data was acquired through a Wheatstone bridge circuit and sampled at 100 Hz. The signal processing was performed in the time response domain. c Springer International Publishing Switzerland 2016 L. Alboul et al. (Eds.): TAROS 2016, LNAI 9716, pp. 288–294, 2016. DOI: 10.1007/978-3-319-40379-3 30
Texture Recognition Using Force Sensitive Resistors
289
In [7], researchers used a metal probe instrumented with an accelerometer and two FSR sensors to classify 69 surface textures “during human freehand movement” with non-constant speed and contact pressure. The accelerometer signal was sampled at 10 kHz, the FSR data was only used to estimate surface friction by measuring lateral forces exerted by the operator’s hand and was not in contact with the surface. Results of two different experiments conducted using a tactile array force sensors attached to a robotic fingertip and moved across the test surfaces in a rubbing motion are presented in [1,2]. The signals were processed using NeuralNetworks and were able to recognise surface textures.
3
Theory
The human sense of touch is achieved through four nerve channels that connect the brain to four types of sensory elements in the skin known as mechanoreceptors [4]. The nerve channels are categorised according to receptor receptive field diameter into type I (receptors with a small receptive field) and type II (rece
Data Loading...
