Novel approach for calculating motion feedback in teleoperation

In this paper we present a method for calculating inertial motion feedback in a teleoperation setup. For this we make a distinction between vehicle-state feedback that depends on the physical motion of the remote vehicle, and task-related motion feedback that provides information about the teleoperation task. By providing motion feedback that is independent of vehicle motion we exploit the spatial decoupling between the operator and the controlled vehicle. In addition, we describe a teleoperation setup that was used to experimentally test the proposed method. In this setup, operators were able to control a remote octorotor from within a control station that is part of a motion simulator, providing visual and inertial feedback. A teleoperation experiment was conducted where 9 participants performed a series of precision hover tasks. The conditions included a no motion condition where only visual feedback was provided, a lateral motion condition, where participants experienced visual motion and the lateral motion of the octorotor, and an error feedback condition, where participants experienced visual motion and the offset between the octorotor and the target as a roll motion. Although we did not find a significant difference in the accumulated error between conditions, results indicate that participants use the motion feedback in their control strategy. After changing the motion feedback condition, the measurements showed a spike in the accumulated error. This spike can be explained by participants using motion feedback in their control strategy and a change in feedback requires a readjustment of this strategy. We see this as a strong indication of motion feedback being helpful for the operator.