Search for: All records

A HRI study with 31 expert robot operators established that an external viewpoint from an assisting robot could increase teleoperation performance by 14% to 58% while reducing human error by 87% to 100% This video illustrates those findings with a side-by-side comparison of the best and worst viewpoints for the passability and traversability affordances. The passability scenario uses a small unmanned aerial system as a visual assistant that can reach any viewpoint on the idealized hemisphere surrounding the task action. The traversability scenario uses a small ground robot that is restricted to a subset of viewpoints that are reachable. 

This paper presents the visual localization subsystem of the ongoing EMILY project. This project aims to assist responders with establishing contact with drowning victims as quickly as possible through the use of a 1.3-meter autonomous unmanned surface vehicle (USV). Once victims are reached, the device can serve as a flotation device to support them. An unmanned aerial vehicle (UAV) provides a live video feed to responders and aids USV's navigation by visually estimating the USV's position and orientation. The movement of the UAV and the USV along with varying lighting, distance, and perspective make the task of estimating USV's position and orientation challenging. We present two implemented solutions for reliable visual localization, the first relying on color thresholding and contour detection, and the second using histograms, back-projection, and CamShift algorithm. The visual localization system was validated in four proof-of-concept field trials. Other unsuccessful methods are discussed.