A novel method based on long short term memory network and discrete-time zeroing neural algorithm for upper-limb continuous estimation using sEMG signals

In this paper, a novel closed-loop model based on surface electromyography (sEMG) comprised long short term memory (LSTM) network and discrete-time zeroing neural algorithm called (ZNN), which is developed to estimate joint angles angular velocities of human upper limb with damping. The dynamic damping set up as the initial equation. Then, LSTM proposed an open-loop described input-output relationship between sEMG signals motion intention. Besides, built via ZNN for eliminating predicted error improving accuracy intention recognition. Founded signals, continuous movement can be successfully estimated model. results show that simple movements, able high accuracy.