Neural network robotic control for a reduced order position/force model

Abstract

The double-support phase is an important walking process to guarantee a smooth switching motion during the locomotion of bipeds.However,the traditional coupled position/force controller can hardly achieve a stable and smooth motion for this phase.A robotic control method is proposed based on a reduced order position/force hybrid robotic model in this paper.The walking locomotion of biped robots in the double-support phase is modeled as a reduced order position/force hybrid model,where the position and force control models are integrated to consider various control performances as a whole and to reduce the complexity of the controller design.The neural network adaptive control method is then pre- sented to guarantee the smooth locomotion and to attenuate the effect of external disturbances and parametric uncertainties. Simulation results are also reported to show the performance of the proposed control model and control scheme.