A model-based approach to recognition of glossy objects is presented. Normals of surface patches are obtained by analyzing the polarizational state of the observed rays under illumination of light sources of circular polarization. The object is assumed to lie alone in its stable pose on the floor. Solid models are examined to find the one which matches the observed normals. First, find a candidate model based on relative angles between known surface normals. Then, find the translation so that observed positions of surface normals coincide with those of models in the image. Some examples are also presented.
In order to realize a mobile robot which moves around on the surface of the wall or the ceiling of building and a large construction, the device which enables to produce strong and reliable adhereing force to the surface must be investigated for the first place. The paper discusses the magnetic device for a mobile robot on the iron wall of nuclear power plant or the ship construction. The comparison of the conventional magnetic device is done first, and then a new magnetic device, named IB magnet, is proposed. It is a device which enables to produce strong magnetic force by parmanent magnet and is easily detachable. The designing methods of the TB magnet are discussed. The feasibility to realize the mobile robot with the TB magnet is demonstrated by a experiment of the man who wears the IB magnet units on each legs and hands and walks on the iron wall and ceiling.