Towards an asynchronous robot control system using the sEMG signals of sequential upper limb movements

• Published in: Chinese Control Conference pp. 5141 - 5145
• Main Authors: Boyang Zhang, Erwei Yin, Jun Jiang, Zongtan Zhou, Dewen Hu
• Format: Conference Proceeding
• Language: English
• Published: Technical Committee on Control Theory, CAA 01.07.2017
• Subjects: asynchronous robot control system; decision-tree; Feature extraction; feature-fusion recognition; Human computer interaction; mode-switched mapping control strategy; Muscles; Process control; Robot control; sEMG signals; sequential upper limb movements
• ISSN: 1934-1768
• DOI: 10.23919/ChiCC.2017.8028167

To date, surface electromyography (sEMG) signals have already been widely and effectively used for the detection and recognition of human limb movements, and applied as control signals in human-computer interaction (HCI) systems. Current trends indicate that taking natural human motion mode as input signals to control peripheral equipments, such as a robot or a mechanical arm, is one of the major subjects in the field of HCI. As previous researches indicated, human complex and continuous motions are composed of sequential simple and discrete sub-movements. In this paper, we proposed and realized an asynchronous robot control system by recognizing the sEMG signals of sequential upper limb movements. Specifically, we designed a mode-switched mapping control system based on seven-class sequences. Furthermore, we extracted the features of mean absolute value (MAV) and waveform length (WL), and used feature-fusion recognition method based on decision-tree. The average control time and actual command number of the subjects reached 110.3 seconds and 17.3, respectively; the best one's indexes reached 89 seconds and 14.7, respectively. The experimental results indicated that the proposed approach could be considered as an idealized model for natural human motions and contributed to realize a more natural HCI system by recognizing upper limb movements sequentially.