Model-based variable position mapping for telerobotic assistance in a cylindrical environment

Inadequacies in sensor and image processing technology have limited the capabilities of autonomous robotics in complex environments, while tedium and fatigue are commonly encountered when purely manual teleoperation is used. Therefore, attention has been increasingly focused on the integration of human and machine control for remote operations. A previously introduced method of assisted teleoperation, consisting of the variable mapping of velocity commands from the master to slave manipulator using uncertain environmental data, is briefly reviewed. The new method is adapted for the task of maneuvering a manipulator within an enclosed cylindrical environment. A model is used to create a mapping from master to slave workspaces to allow the operator to move around the curved surface easily while avoiding unintended impacts with the wall. Experiments were developed which required fine positioning of the end-effector near the wall, and the results demonstrated the increase in efficiency of task completion and reduction of operator fatigue and skill requirements