Dual arm movement control by a neurobotics model

One of the amazing successes of biological systems is the ability of animals to learn to control the complicated dynamics of their muscles and joints smoothly and efficiently. Traditional engineering control techniques, on the other hand, often do not perform well when confronted with intrinsically complex systems with many degrees of freedom such as a robot arm (human arm). This paper presents new work on compliant motion control. Based on previous physiological information, the authors propose an intelligent adaptive system based on a decentralised motor learning model of the cerebellum. The resultant neuro-adaptive model is used as a hybrid force/position controller for a dual arm. To optimise the neural network learning strategy, a hybrid neuro-genetic algorithm is introduced and simulation results are given for comparisons.