Home Projects Publications Teaching Resume

Evaluation of Gesture-Based Controls for Robotic Systems

Mubbasir Kapadia, Lauren Frazier, Norman I. Badler

Abstract

Robotic control systems are becoming more common, especially in the military, and there is a growing need for better control interfaces. Arm and hand gestures are typical human forms of communication, so applying that to a robotic control system can yield a more intuitive system. With military applications, there are lives at stake, so having the most efficient, intuitive control system can make a significant difference in the success of a mission and the safety of the soldiers involved. We present a comparative analysis of 4 different controllers on a robot navigation task. A commercial smartphone platform is used to provide control using gestures, a D-pad, and a tilt-based control scheme, whereas an XBOX 360 controller is used to provide traditional joypad-based control. We perform a human factors experiment to control a Roomba robot on a navigation task and collect quantitative as well as qualitative metrics to evaluate the quality of the controllers. Results indicate that gesture-based control is not favorable when mapped to low-level navigation directives, suggesting the exploration of a high-level command suite for gesture-based robot control.

--> myspace visitor tracker