Continuous Motion Estimation of Lower Limb Joints Based on BP-KPCA Multi-feature Fusion.

In order to realize the accurate and stable prediction of lower limb knee joint continuous motion with different motion modes, a lower limb continuous motion prediction model based on the combination of kernel principal component analysis (KPCA) and BP (back propagation) neural network is proposed in this paper. Firstly, the time-domain characteristics of the preselected EMG signals of four muscles are extracted. Secondly, they are fused to remove the redundancy by KPCA and get several principal components with complementary information. Finally, four representative principal components are selected as inputs and imported into the prediction model. The experimental results show that the method can effectively predict the joint angle of lower limb movements.