Development of the Human Interactive Autonomy for the Shared Teleoperation of Mobile Robots

Manipulation complexity of teleoperation increases the necessity of shared autonomy. However, designing an autonomy while keeping in mind the environmental uncertainty is challenging. This paper proposes a human interactive autonomy for a shared teleoperation. The central concept is a novel methodology of combining the innate cognitive ability of a human operator to arrive at an accurate, and precise solution of an autonomous system for mobile robot shared teleoperation. The resulting system is more efficient and less fatigued than direct teleoperation. This combined approach transfers the human operator's intention, shaped by his cognition about remote environment, to the autonomous system. The intention information input from the operator helps autonomous system perform the given teleoperation tasks more efficiently. For initial feasibility and proof of concept, the intention information is provided to the autonomous system in the form of a path which is acquired by our proposed sketch method. Our novel proposed sketch method allows the operator to sketch the path on the visual feedback image of the remote environments and the information is then transmitted to the autonomous system. Experimental results provide the feasibility and effectiveness of proposed method against prevalent control methods.