Advanced Methods of Human Machine Interface for Improved Myoelectric Control

点击翻译按钮获取中文摘要

While prosthesis design has evolved greatly over the last several decades, current artificial limbs do not fully replace the intricate, multifunctional human limb. Advancements in materials and technology have led to improved human¬ machine interfaces, but operations can be slow and are often not intuitive, leading to user dissatisfaction and abandonment of the device. Traditionally, surface electromyography (EMG) is used as the control input for powered upper limb prostheses. The advent of multichannel signal recordings or high density EMG (HDEMG) is a promising technique that collects signals from many closely spaced electrodes, resulting in accurate and detailed muscle activation data. Harnessing the information contained within HDEMG will advance research towards the development of a dexterous upper limb prosthesis with multifunction control. Combining HDEMG data with neural actitivty measure through electroencaephalography (EEG) will further advance knowledge to examine cognitive burden on the device user. The ability to combine these biological signals will advance efforts to develop artificial limbs. The proposed research program will be conducted in the Andrew and Marjorie McCain Human Performance Laboratory (HPL) at the University of New Brunswick. The HPL is a state¬-of¬-the-¬art research, education and clinical service facility within the Faculty of Kinesiology at UNB. The HPL is well equipped with highly specialized equipment to study human movement and has significant experience in prosthesis design, control and development. This research program aims to develop more intuitive myoelectric control systems that are capable of many degrees of freedom to more closely resemble intact human limbs. Through the use of biological sensors including HDEMG and EEG we will explore muscular and neural components of muscle activation and signal control to develop new methods of myoelectric control. Analysis of the data will require advanced signal processing techniques as well as the development of new algorithms to investigate muscle behavior and neuromuscular function. We will also exploit the tremendous recent increase in availability of wireless sensor technology to develop new and innovative systems of obtaining muscle data (for example e¬textiles). This research has applications in a number of areas of biomedical engineering from prosthesis design to development of exoskeletons. Over the next five years, the proposed research program will support four graduate students and four undergraduate students and provide them with significant training in state¬-of-¬the-¬art myoelectric control systems and prosthesis design.