Development of a forward-hemispherical vision sensor for acquisition of a panoramic integration map

Abstract

The target of this research is to develop a common-use sensor which is useful for navigation and mission for rescue operation. By combined use of an ODV (omnidirectional vision) with a hemispherical forward-visual-field using a direct and reflection hybrid optical system, a RTK-GPS (real time kinematics-global positioning system) and a MEMS (micro electro mechanical systems) IMU (inertial measurement unit), this sensor outputs a realtime image for remote control of a mobile robot and a panoramic integration map, including information on three-dimensional own-position, with a continuous panoramic image for searching victims and for making a rescue plan. The evaluation of the panoramic integration map and the Kalman filter based self-positioning reveals that the proposed method is valid and effective