Neural network control of VietCobot

Abstract

This paper is a research to come of our developing cobot, VietCobot, for service and medical applications. In this paper a neural network control is applied to a six degrees of freedom collaborative robot to observe the system dynamics behavior which play an important role in force control. To do so, the dynamics equation of the cobot is derived taking into account parametric uncertainties and external disturbances; The forward kinematics and reverse dynamics of the cobot is solved; then, a neural network controller is derived to compensate the uncertainty and disturbances based on the dynamics desired trajectory and sensory data of the joints. The simulation results are included to illustrate the performance of the neural network control strategy. In addition, an all-in-one cobot integrated development environment (cobot IDE) software is developed to integrate the controller simulation and the experimental works for a complete research tool.