To restore function for paraplegic patients, Functional Electrical Stimulation (FES) is an effective method. FES can realize some Activities of Daily Livings (ADL) such as standing-up, ambulation and grasping. However there are several unsettled problems that prevent FES from further practical uses. One is about its switch. FES stimulation system needs a switch signal to denote the onset of stimulation. However in most of the current FIES technology, patients have to make a superfluous action by themselves or rely on someone else to turn on-off the switch instead. To release patients from such a switching action, we have been developing another type of switch for FES control for hemiplegic patients, based on the consideration that, lower limbs' activities, for example, standing-up and ambulation, need the synchronization of both limbs. So that, in this case, it is possible to recognize the activity that the patients intend to do, from some kind of physiological signal detected from their normal side limb. In this research, we use EMG (electromyogram) as the signal. EMG is physiological signal with individual disparity, which inevitably results in classification error. To overcome this problem, we proposed an ANN based online classifier system incorporated with a problem-oriented feature extraction method. This paper will present such an adaptive automatic switch system.
展开▼
