Analysis of the Leap Motion Controller Workspace for HRI Gesture Applications

Abstract

The Leap Motion Controller is a sensor for precise hand tracking; it is a device used for human interaction with computer systems via gestures. The study presented in this paper evaluates its workspace in real-world conditions. An exact replica of a human operator’s hand was used to measure the sensor’s precision, and therefore determine its hand tracking abilities in varying positions above the sensor. The replica was moved randomly across the workspace defined by the manufacturer, and precision was measured in each position. The hand model was placed in the furthest distances from the sensor to find every position where the sensor was still capable of tracking. We found the dimensions of the workspace in some cases exceeded the datasheet values; in other cases, the real workspace was smaller than the proclaimed one. We also computed precision in all positions, which shows tracking reliability. This study serves researchers developing HMI and HRI algorithms as a reference for the real dimensions of the Leap Motion Controller workspace as it provides extra and more precise information compared to the datasheet.

Keywords

• leap motion controller
• workspace evaluation
• HMI (human–machine interaction)
• HRI (human–robot interaction)
• UR5
• hand gesture recognition