Towards an EMG-Controlled Prosthetic Hand Using a 3-D Electromagnetic Positioning System

Abstract

This paper proposes a novel method of using electromyographic (EMG) potentials generated by the forearm muscles during hand and finger movements to control an artificial prosthetic hand worn by an amputee. Surface EMG sensors were used to record a sequence of forearm EMG potential signals via a PC sound card and a novel 3-D electromagnetic positioning system together with a data-glove mounted with 11 miniature electromagnetic sensors used to acquire corresponding human hand pose in real time. The synchronized measurements of hand posture and associated EMG signals stored as prototypes embody a numerical expression of the current hand shape in the form of a series of data frames, each comprising a set of postures and associated EMG data. This allows a computer generated graphical 3-D model, combined with synthesized EMG signals, to be used to evaluate the approach. This graphical user interface could also enable handicapped users to practice controlling a robotic prosthetic hand using EMG signals derived from their forearm muscles. We believe this task might be made easier using a dictionary of stored task-specific prototype data frames acquired from able-bodied users. By comparing the resulting EMG data frames with stored prototypes, the most likely data frame sequence can be identified and used to control a robotic hand so that it carries out the user's desire. We explore the feasibility of this approach by applying frequency analysis on the signal derived from a multichannel EMG measurement device and identify pattern recognition techniques in the time and frequency domains to determine plausible hand shapes. This approach offers several advantages over existing methods. First, it simplifies the classification procedure, saving computational time and the requirement for the optimization process, and second, it increases the number of recognizable hand shapes, which in turn improves the dexterity of the prosthetic hand and the quality of life for amputees. The database of EMG prototypes could be employed to optimize the accuracy of the system within a machine learning paradigm. By making a range of EMG prototype databases available, prosthetic hand users could train themselves to use their prosthesis using the visual reference afforded by the virtual hand model to provide feedback