The gestural joystick and the efficacy of the path tortuosity metric for human/robot interaction

Abstract

This paper describes the design and evaluation of the "gestural joystick," a wearable 2-D pointing controller for mobile robots in hazardous environments that uses hand gestures. Hazardous environments, such as that of a collapsed building search, require operators to wear a significant amount of protective clothing. This protective clothing, which may include hard hats, suits, gloves, goggles, etc., reduces comfort, mobility, dexterity, load capacity, and ability to interact with conventional computer input devices. The gestural joystick, which is embedded in protective clothing, mitigates some of these impacts, but at the cost of lesser familiarity for the user and, therefore, potentially lesser performance. Effective performance metrics are required to evaluate this interface mechanism. Path tortuosity has been proposed as a performance metric for the evaluation of teleoperation of a robot, but has not been proven to be distinct from time-to-complete metrics. By injecting controlled uncertainty between the user and robot, we show, for the first time, that path tortuosity is a useful and distinct metric for the evaluation of robot teleoperation.