Advanced biped gait generator using NARX-MLP neural model optimized by enhanced evolutionary algorithm

A novel biped walking pattern combining robust zero-moment-point ZMP technique and pre-determined foot-lifting value is proposed in this paper. The implementation of suggested approach contains following stages. Initially, a one-step curve for small-sized humanoid created using the 3rd-order interpolating equation, with velocity responding concept. next step, gait planning modeled as non-linear MIMO plant including ten degree-of-freedom DOF. Then, installation generator (WPG) based on new hybrid Neural-NARX model completed. Eventually, Enhanced Differential Evolution (EDE) applied to optimally identify weights structure, ensuring robot terms desired trajectories value. All case studies confirm that it surely provide WPG satisfying both effectiveness high robustness. verification newly adequately tested via simulation experiment results.