Model-based and neural-network-based adaptive control of two robotic arms manipulating an object with relative motion

Abstract

In the study of constrained multiple robot control, the relative motion between the constraint object and the end effectors of manipulators are usually neglected in the literature. However, in many industrial applications, such as assembly and machining, the constraint object is required to move with respect to not only the world coordinates but also the end effectors of the robotic arms. In this paper, dynamic modelling of two robotic arms manipulating an object with relative motion is presented first, then a model-based adaptive controller and a model-free neural network controller are developed. Both controllers guarantee the asymptotic tracking of the constraint object and the boundedness of the constraint force. Asymptotic convergence of the constraint force can also be achieved under certain conditions. Simulation studies are conducted to verify the effectiveness of the approaches.