Development of a laser range finder for 3D map-building in rubble

Abstract

In order to search and rescue the victims in rubble effectively, a 3D map of the rubble is required. As a part of the national project of rescue robot system, we are investigating a method for constructing a 3D map of rubble by teleoperated mobile robots. In previous work, we have developed the first prototype of laser range finder for rubble measurement. The developed one consists of a ring laser beam module and an omnivision camera. However, the first prototype has the weakness in which the measurement can be performed only in a dark place, because it is difficult to extract the reflected laser image from the picture captured in a light place. On the contrary, the pictures, in which the contrast of brightness is clarified, are required in order to extract the environmental features which are utilized for the position estimation of the robot based on the SLAM frame work. Also, the texture images captured in a light place are required for constructing virtualized rubble. In this paper, the second prototype of laser range finder which we developed in order to solve above dilemma is described. Although the concept of this prototype is similar to the first one, it has an infrared laser module and two cameras for capturing the reflected laser image and texture image separately. And, a hot mirror is used to dissociate the infrared laser ray from visible rays. We evaluate the measurement accuracy of the second prototype based on the measurement error analysis described in previous work. Using the second laser range finder, 3D maps of rubble were actually built with reasonable accuracy.

Improving Search and Rescue Using Contextual Information

“…The main disadvantage of using these systems for robotic 3-D mapping of USAR environments is that they are slow and require a lot of time for scanning, due to the fact that the laser stripe has to be physically moved across the scene to digitize the surface and, hence, they cannot provide real-time range data acquisition. Alternatively, Kurisu et al developed two different laser range finders for 3-D mapping of rubble in USAR environments [15]. The first laser range finder consists of a ring of laser beam modules and an omnivision CCD camera.…”

3D panorama stereo visual perception centering on the observers

“…The iterative closest point (ICP) is able to splice two scanning stations, but errors may be present in any kind of splicing and it costs large computing resources [5]. To address the problems of the traditional laser scanning techniques, researchers have proposed various active stereo vision technology-based solutions [3,4,6,8,[14][15][16].…”

“…This method does not require mechanical parts and can spread the stripes over larger areas. Recently, Kurisu et al [4] proposed a 3D map construction technique using two cameras, circle laser generators, and conic mirror. Matsui et al [6] detected the panorama of piping using an omni-directional camera and an omni-directional laser.…”