Stable force-myographic control of a prosthetic hand using incremental learning

摘要: Force myography has been proposed as an appealing alternative to electromyography for control of upper limb prosthesis. A limitation this technique is the non-stationary nature recorded force data. patterns vary under influence various factors such change in orientation and position We hereby propose incremental learning method overcome limitation. use online sequential extreme machine where occasional updates allow continual adaptation signal changes. The applicability effectiveness approach demonstrated predicting hand status from forearm muscle forces at arm positions. results show that are indeed effective maintain a stable level performance, achieving average classification accuracy 98.75% two subjects.