Acceleration consensus for networked motion control of telerobots

Abstract

Consensus algorithms basically enable nodes on a graph to dynamically agree on certain values. In this research, a second order consensus algorithm with a servo input, similar to an acceleration controller, is devised; making it possible for robots on a network to agree on their acceleration and track the average of local servo inputs. The agreement on acceleration by the robots employing consensus algorithms is called acceleration consensus. Furthermore it is also shown that teleoperation can be described as a consensus problem and therefore this problem can be solved with consensus algorithms. If the servo inputs of an acceleration consensus algorithm are written as a function of the forces acting on each robot, we can realize stable haptic teleoperation among any number of robots. The experiments verify that acceleration consensus can be realized on various network topologies, enabling haptic teleoperation with multiple teleoperators.