Assigned responsibility for remote robot operation

The remote control of robots, known as teleoperation, is a non-trivial task, requiring the operator to make decisions based on the information relayed by the robot about its own status as well as its surroundings. This places the operator under significant cognitive load. A solution to this involves sharing this load between the human operator and automated operators. This paper builds on the idea of adjustable autonomy, proposing Assigned Responsibility, a way of clearly delimiting control responsibility over one or more robots between human and automated operators. An architecture for implementing Assigned Responsibility is presented.