Development of Robot Teleoperation System in Bad Viewing Condition

Abstract

In robot teleoperation there is a mounted camera on a robot and the operation is usually performed from a remote site using captured images by the mounted camera. Even though color cameras provide many useful information of a remote site, robot teleoperation using color cameras are highly effected by environmental conditions such as lighting, colors, smoke, etc. and there might be a case that cameras might become useless during an operation such as in dark places. In this paper we have developed a robot teleoperation system which does not fully rely on color camera images and works well in bad viewing condition for operating a robot and searching for a victim. A laser rangefinder is used for sensing surroundings of a robot and a thermal camera is used for victim detection. An operator can control a robot in a remote site only using environmental contour figure information around a robot without color camera images. The augmented image interface consists of color and thermal camera images are used for victim detection. We have executed a victim searching task in different lighting conditions with different sensor configurations to show the effectiveness of developed system.