Recognition is an important function for robots which supports human in life environment and searching at disaster sites. When the robots decide the next motion, the robots need to sense the surrounding environment and recognize them. Good recognition can allow the robot to decide good motion. In this research, we aim to develop methods of various objects recognition and to create digital model of the world.


Partner robots, who can co-operate with humans, have to know the various information of objects in daily-life environment. However the information of objects in real environment are unknown and it is needed to acquire . In PREST project, we aim to gather fine information of unknown objects and to create database for the autonomous robots. We propose a method of unknown object recognition based on visual and motion clues.

3-D Laser Scanner and 3-D Mapping

We have developed a method of 3-D mapping by using a tracked vehicle with 3-D laser scanner. We proposed a reticulate scan method for measuring dense 3-D shape. 3-D shapes are measured at several locations as point cloud data by using the reticulate scan method during exploration and 3-D map is built by matching these 3-D point cloud data with each other. We have evaluated and improved the 3-D mapping method with Japanese firefighters.