Cited time in webofscience Cited time in scopus

Tactile Signal Processing and Haptic Interface System

Title
Tactile Signal Processing and Haptic Interface System
Alternative Title
촉각 신호 처리 및 햅틱 인터페이스 시스템
Author(s)
Kwonsik Shin
DGIST Authors
Kwonsik ShinJae Eun JangHongsoo Choi
Advisor
장재은
Co-Advisor(s)
Hongsoo Choi
Issued Date
2022
Awarded Date
2022/08
Type
Thesis
Subject
Tactile sensor, Tactile actuator, Piezoelectricity, Electrical stimulation, Tactile signal processing, Haptic interface system
Description
Tactile sensor, Tactile actuator, Piezoelectricity, Electrical stimulation, Tactile signal processing, Haptic interface system
Abstract
요즈음 촉각에 대한 연구에 관심이 높아지면서 햅틱 인터페이스 시스템도 많이 발전되고 있다. 최근에 많은 사람들이 스마트폰, 터치 디스플레이 및 VR 장비를 통해 햅틱 기술을 사용하고 있으며, 다양한 유형의 장치들이 감도, 효율성, 신뢰성 및 유연성과 같은 성능 향상을 위해 개발되고 있다. 예를 들어 기존의 실리콘 기반 센서에서 벗어나 최근에는 웨어러블 하고 플랙서블한 햅틴 센서뿐만 아니라 사람의 피부를 모방한 E-skin 디바이스가 개발되는 중이며, 사람의 촉각 시스템보다 더 민감하고 고해상도 성능을 갖는 다양한 햅틱 장치 또한 다양하게 연구되고 있는 중이다. 햅틱 인터페이스 시스템은 일반적으로 다양한 자극을 감지하는 촉각 센서와 햅틱 피드백을 사용하여 인공 촉각 감각을 재현하는 촉각 액추에이터로 구성되어 있으며 센서와 액추에이터의 발전덕분에 최근에는 수술용 로봇 팔, 재활 기기 등에도 햅틱 시스템이 활용되고 있다. 그러나 이러한 기술의 발전에도 불구하고, 사람은 압력이나 온도와 같은 물리적 자극뿐만 아니라 거칠기, 딱딱함, 심지어는 고통과 같은 정신감각적 자극 느끼기 때문에, 인간의 촉각 시스템을 완벽하게 모방하는 데에는 많은 과제가 남아 있다. 게다가 사람마다 이러한 정신감각적인 느낌을 감지하는 기준이 다르기 때문에 햅틱 피드백을 통해 인공적으로 촉감을 재현하는 것 또한 많은 이슈가 있다. 따라서 기존의 햅틱 시스템이 가지는 한계를 뛰어넘어 물리적, 정신감각적 감정을 모두 다룰 수 있는 첨단 햅틱 인터페이스 시스템에 대한 연구가 필요하다.
본 논문은 압전 촉각 센서와 전기 자극 액추에이터를 이용한 첨단 햅틱 인터페이스 시스템의 개발에 중점을 두고 있다. 압전 효과는 빠른 응답속도와 높은 민감도를 가지기 때문에 정적뿐만 아니라 동적 모션을 감지하고 센싱하는데 매우 적합합니다. 또한 압전효과는 외부에서 가해진 물리적인 스트레스로부터 스스로 전기를 만들 수 있는 독특한 특징을 가지고 있어서, 시스템을 간략화 할 수 있으며 추가적인 외부 전원이 불 필요하다는 장점이 있다. 햅틱 피드백을 위한 전기자극 액추에이터는 주파수, 신호파형, 진폭 등 여러 변수를 컨트롤 할 수 있어서 사람에게 인공 촉감을 재현할 수 있는 효과적이고 효율적인 방법 중 하나이다. 또한 SA 및 FA 촉각 수용체의 메커니즘을 활용하면 다양한 인공 촉각을 재현할 수 있을 것으로 기대된다.
첨단 햅틱 인터페이스 시스템 개발을 위하여 먼저 표면 정보를 측정할 수 있는 압전 소재 기반의 유연 촉각 센서를 연구를 진행했다. 압전 유연 센서는 고 해상도와 높은 정확도를 위하여 어레이 구조로 디자인되었으며, 센서에서 발생되는 압전 신호를 분석하여 표면 정보를 추출하는 신호처리 기술을 연구하였다. 표면이 가지는 특정한 패턴, 모양 및 변위와 같은 측정 가능한 물리적 인자들을 효과적으로 측정하기 위하여 슬라이딩 모션을 사용했으며, 동적인 모션을 사용했기에 측정되는 압전신호들은 시간 도메인에서 분석되었다. 개발된 압전 센서는 높은 선형성과 우수한 신뢰성 및 안정적인 감도를 가졌으며, 200μmm 넓이와, 500μm의 미세한 패턴을 측정할 수 있음을 확인했다.
두 번째로, 기존의 압전 센서를 핀 형태의 모듈과 결합하여 깊이 정보를 측정할 수 있도록 개선하는 연구를 진행했다. 핀 형태의 모듈은 스프링 및 벌크한 구조를 가지지만, 3차원 정보 측정의 한계를 극복할 수 있는 센싱 메커니즘을 제안한다. 또한, 압전 신호로 부터 깊이 정보를 추출할 수 있는 신호 처리를 개발하여, 표면에 있는 다양한 모양, 간격, 접촉각도 등을 측정 후 신호 처리하여 표면 정보를 이미지로 렌더링할 수 있다. 이러한 결과를 기반으로 하여, 직물과 같은 부드러운 재료의 표면 정보 또한 3D 이미지화 할 수 있었다.
또한 슬라이딩 조건에서 압전 신호에 대한 근본적이고 기초적인 분석을 진행하였다. 기존의 압전 신호 분석은 주로 피크 출력 전압과 관련이 있기 때문에 동적 상황에 해당하는 시간 경과에 따라 압전 신호를 분석할 때 많은 이슈와 한계가 있었다. 이를 극복하기 위하여 작은 세그먼트와 유닛 전극에서 발생되는 압전 신호를 비교하여 압전 신호 발생 과정에 대한 단서를 얻고 압전신호를 수학적으로 분석하여 일반해를 구하였다. 압전 신호를 익스포넨셜 및 오차 함수로 표모델링 된 압전 신호를 기반으로 인공 압전 신호를 만들었고, 이러한 인공 신호와 실제 압전 신호를 비교하여 일반해를 최적화하여 인공 압전신호의 신뢰성과 재현성을 향상시켰다. 만들어진 인공 신호들을 조합에서 다양한 깊이 프로파일들을 측정하였다.
마지막으로 전기 자극 메커니즘을 이용하여 인공 촉각 재생을 위한 촉각 액츄에이터를 개발하였다. 제안된 액추에이터는 PCB 기반의 어레이 전극을 가지며 전극의 크기, 전극의 피치 및 접지 구조와 같은 구조상의 변수들을 고려하여 전기 자극을 최적화한다. 또한 Labview 프로그램으로 전기자극에 사용될 다양한 전기 신호를 개발하고 주파수, 신호 파형 및 진폭을 제어하여 커스텀 및 최적화를 진행하였다. 다양한 변수들을 조절한 결과를 바탕으로, 참가자들의 공간적 해상도 향상을 확인했으며, 이러한 방식을 최적화한다면 다양한 자극을 재현할 수 있을 것으로 기대된다.
|The study of the human tactile sense is getting interested, along with the development of haptic technology. Haptic technology is one of the advanced interactions between humans and devices, so it has been developed for bilateral communication and surgical robot. Therefore, many types of sensor mechanisms and advanced materials have been studied to mimic the characteristics and responses of human tactile system. With the development of tactile sensors, tactile actuators have also been introduced to reproduce artificial tactile sensations. As a result of the development of tactile sensors and actuators, haptic interface systems with characteristics of sensor and actuator are studied to not only sense various tactile stimuli but also feedback realistic tactile sensations in field of VR/AR devices, and rehabilitation devices. However, despite advances in these technologies, there remain numerous challenges to perfectly mimicking the human tactile system because it is not easy to sense psychological sensations such as roughness, hardness, and texture. In addition, it is not easy to create artificial tactile feedback because people have different criteria for sensing these psychological feelings. Therefore, it is necessary to make multimodal tactile sensors that can sense various surface properties, and advanced tactile actuators which can create realistic tactile feedback for a haptic interface system.
This thesis focuses on developing an advanced haptic interface system with a piezoelectric tactile sensor and electrical stimulation actuator. The piezoelectric effect offers rapid responses and high sensitivity, so that it is appropriate to detect static and dynamic motions. In addition, piezoelectricity can generate electrical signals in response to applied stress, taking advantage of the self-powered characteristics. A sliding motion of sensor was studied mainly here because various surface information such as pattern, shape, and displacement can be sensed rather than simple touch motion. Besides, a specific signal processing was suggested and applied to surface topography reproduction. Electrical stimulation was studied for tactile actuators because this method can make artificial tactile feelings without any touch motions by adjusting frequency, signal waveform, and amplitude. Therefore, it is expected to be utilized for realistic artificial tactile feedback.
As the first approach for developing the advanced haptic interface system, a tactile array sensor with piezoelectric material was studied to measure surface topography. Array sensor structure and signal analysis skills were proposed to measure surface topography with high accuracy. The suggested array sensor showed high linearity (R^2=0.98), good reliability (250 times), and stable sensitivity (0.1V/N), as well as could measure a contact width and pitch information based on the measured piezoelectric voltage signals. Additionally, the piezoelectric sensor could calculate an angle of diagonal sliding motion. It was confirmed that the sensor could get x and y-axis information with high accuracy in any sliding direction. However, because of the limit of flexibility in senor structure which resulted in low resolution of the z-axis, 3D surface topography was not reproduced perfectly. Therefore, additional treatment is required for high sensing performance.
In order to overcome the issue and to improve sensing performance in z-axis information, a piezoelectric sensor e with a pin-module was proposed. The sensor combining a pin-type module showed excellent monitoring in the depth direction by pin and spring motion, so that it could have an excellent resolution of the depth information. The depth information was calculated by integrated piezoelectric voltage signals because the integral values of the piezoelectric signal have a linear relationship with various depth conditions. The integrated piezoelectric signals were used for measuring contact width, spacing, and contact angle. Therefore, surface topography was reproduced with high accuracy. As one of real applications, the surface information of soft fabric sample was measured with high accuracy by the sensor.
Additionally, we demonstrated a fundamental analysis of a piezoelectric signal in a sliding condition. Because a conventional analysis of piezoelectric signals is mainly related to the peak output voltage, it has limits of sensing resolution for time domain. Therefore, piezoelectric signals were analyzed in the time domain corresponding to dynamic situations. We compared the piezoelectric signals between small segments and a unit cell to obtain a clue about the signal process and confirm the mathematical formula. Based on the results of piezoelectric signal fitting with the exponential and error function, general solutions that can model piezoelectric signals are proposed to create artificial piezoelectric signals. By comparing the artificial and measured signals, the general solution is optimized, and the created artificial piezoelectric signals are found to have good reliability and reproducibility. Various depth profiles can be calculated from the combination of the artificial signals. In addition, the simultaneously measurable depth profiles are improved with more artificial signals. When tactile sensors can deform or bend well on a touching object, the signal process can estimate multiple depth profiles of the object.
With the development of a tactile sensor, a tactile actuator for artificial tactile sensation was studied by using electrical stimulation mechanism. The proposed actuator had an array of electrodes for stimulation. Several design parameters such as the size of the electrode, the pitch of electrodes, and ground structure were considered to optimize electrical stimulation. Various voltage signals were developed and customized by controlling a frequency, a signal waveform, and an amplitude. For electrical signal optimization, signal waveforms (sine, square, triangle, and sawtooth) were compared, and effective frequency ranges were also analyzed. Then, spatial resolution was analyzed by varying stimulated areas with human perception test. Most of the participants had a higher resolution sensation in the case of a complex signal using various frequencies and waveforms with small area stimulation. Therefore, it was revealed that composite signals and stimulation areas were important major parameters to reproduce various tactile sensations based on electrical signals.
This thesis introduced a tactile sensor based on the piezoelectric effect and a tactile actuator with electrical stimulation for a haptic interface system. The proposed piezoelectric sensor showed an excellent sensing performance with x and y-axis information. A structural improvement method with pin-module and fundamental signal analysis were proposed to analyze depth profiles. In addition, a tactile actuator system with electrical stimulation was introduced, and electrical signals were optimized by adjusting signal waveforms, frequencies, and stimulation areas. Although there are still challenges that need to be improved, the results of this thesis are expected to be a fundamental study of a multi-functional system that is able to deal with both physical and psychological sensations for an advanced haptic interface.
Table Of Contents
Ⅰ. INTRODUCTION 1
1.1 Haptic Interface System 1
1.2 Human Tactile System 3
1.2.1 Tactile Receptors 5
1.2.2 Tactile Signals 9
1.3 Basic Principle of Tactile Sensor 10
1.3.1 Piezoresistive Tactile Sensor 11
1.3.2 Capacitive Tactile Sensor 13
1.3.3 Triboelectric Tactile Sensor 14
1.3.4 Piezoelectric Tactile Sensor 15
1.4 Basic Principle of Tactile Actuator 16
1.4.1 Kinesthetic Feedback 19
1.4.2 Vibrotactile Feedback 20
1.4.3 Electrical Feedback 21
II. BACKGROUND OF PIEZOELECTRIC EFFECT AND ELECTRICAL STIMULATION 23
2.1 Background 23
2.2 Background of Piezoelectric Effect 25
2.3 Previous Works of Piezoelectric Tactile Sensor 27
2.4 Background of Electrical Stimulation 32
2.5 Previous Works of Electrical Stimulation 34
2.6 Motivation 35
III. THE SENSING PERFORMANCE AND CHARACTERISTICS OF PIEZOELECTRIC SENSOR WITH SLIDING MOTION 39
3.1 Introduction 39
3.2 Sensor Design and fabrication 40
3.3 Basic Performance and Characteristics 42
3.4 Analysis of Piezoelectric Signals with Sliding motion 45
3.5 Surface Topography Measurement 48
IV. IMPROVEMENT OF SENSING CAPABILITY AND 3D STRUCTURE ANALYSIS WITH PIN-MODULES 52
4.1 Introduction 52
4.2 Sensor Design and Fabrication 53
4.3 Analysis of Piezo-Signal for Depth Measurement 57
4.4 Complex Structure Rendering based on Signal Processing 62
4.5 3D-rendering Result of Fabric Samples 67
V. FUNDAMENTAL ANALYSIS OF THE ELECTRICAL SIGNALS OF PIEZOELECTRIC MATERIALS 71
5.1 Introduction 71
5.2 Concept and Simulation 73
5.3 Electrode Design and Fabrication 76
5.4 Signal Comparison between the Unit and Segment Electrodes 77
5.5 Signals processing and General Solution of Piezoelectric signal 83
5.6 Depth Profile Analysis with Artificial Piezoelectric Signals 88
VI. A STUDY OF TACTILE ACTUATOR AND TACTILE SIGNAL PROCESSING BASED ON ELECTRICAL STIMULATION 94
6.1 Introduction 94
6.2 Tactile Actuator System 95
6.3 Tactile Signal Processing with Voltage Mode 98
6.4 Resolution analysis of Electrical Tactile Sensation 101
VII. CONCLUSION 115
VIII. REFERENCES 118
URI
http://dgist.dcollection.net/common/orgView/200000627989

http://hdl.handle.net/20.500.11750/16777
DOI
10.22677/thesis.200000627989
Degree
Doctor
Department
Department of Electrical Engineering and Computer Science
Publisher
DGIST
Related Researcher
  • 장재은 Jang, Jae Eun
  • Research Interests Nanoelectroinc device; 생체 신호 센싱 시스템 및 생체 모방 디바이스; 나노 통신 디바이스
Files in This Item:

There are no files associated with this item.

Appears in Collections:
Department of Electrical Engineering and Computer Science Theses Ph.D.

qrcode

  • twitter
  • facebook
  • mendeley

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE