학술논문
근전도 센서 설계와 근전도 신호를 이용한 힘에 독립적인 인간 팔 동작 추정 알고리즘 / EMG Sensor Design and Human Arm Motion Estimation Algorithm Independented on Force using EMG signal
Document Type
Dissertation/ Thesis
Author
Source
Subject
Language
Korean
Abstract
최근 들어 생체신호 연구에 대한 관심이 높아지고 있다. 그 중에서도 근육의 움직임을 측정하는 EMG 신호는 인간의 동작을 분석하는데 있어서 가장 중요한 신호라 할 수 있다. 인간이 팔을 움직이거나 힘을 주었을 경우 팔에 있는 근육들은 활성화 되기 시작한다. 실제로 EMG 신호는 관절의 동작이나 힘에 의하여 발생되어 진다. 이러한 이유로 EMG 신호를 분석하는 것은 동작정보를 알아내는데 주로 연구되어 왔다. 하지만 힘에 따라 EMG 신호가 변하는 관계로 힘에 따른 관절 각도의 추정은 연구가 어려운 실정이다.이에 본 논문은 EMG 신호를 측정하기 위한 센서 설계 과정을 기술하고, 표면 근전도 센서를 이용하여 Biceps brachii, Deltoid 그리고 Triceps brachii 의 세 부분의 근육에서 획득한 원신호를 Normalization을 통한 기하학적 방법론을 제시하여 힘에 독립적인 관절 각도 추정 알고리즘을 기술하고 실험을 통하여 이를 입증한다.
Recently, It has been increasing interest in research of bio-signal processing method. Among the bio-signals, The EMG signals of muscles are very important ones to analyze how the body of human intends to move. Both motion and force of the human arm produce the activation of several muscles. Actually EMG signal is generated by motion of either joint or force. So the method of EMG signal processing has been studied for analysis of acquisition of motion information. But, It is difficult to estimate the user’s arm motion according to force.In this paper, we describe the Hardware Design of EMG Sensor for acquisition of EMG signal and then we have acquired EMG signals from biceps brachii, deltoid and triceps brachii by using surface EMG sensor board. Secondly, we propose a new method to estimate the user’s arm motion independently of muscles force using normalization and sphere equation. The advantage of this method is to estimate with real-time implementation about biological activity of the muscles independent on force.Lastly, this paper shows the effectiveness of the suggested method through experimental results.
Recently, It has been increasing interest in research of bio-signal processing method. Among the bio-signals, The EMG signals of muscles are very important ones to analyze how the body of human intends to move. Both motion and force of the human arm produce the activation of several muscles. Actually EMG signal is generated by motion of either joint or force. So the method of EMG signal processing has been studied for analysis of acquisition of motion information. But, It is difficult to estimate the user’s arm motion according to force.In this paper, we describe the Hardware Design of EMG Sensor for acquisition of EMG signal and then we have acquired EMG signals from biceps brachii, deltoid and triceps brachii by using surface EMG sensor board. Secondly, we propose a new method to estimate the user’s arm motion independently of muscles force using normalization and sphere equation. The advantage of this method is to estimate with real-time implementation about biological activity of the muscles independent on force.Lastly, this paper shows the effectiveness of the suggested method through experimental results.