|
Accession number;05A0661050
|
| Title;Translating topological maps between heterogeneous robots |
| Author;
KANAZAKI HIROFUMI
(Univ. Tokyo, Graduate School of Engineering, JPN)
HORI KOICHI
(Univ. Tokyo, Graduate School of Engineering, JPN)
|
Journal Title;Proceedings of the Annual Conference on JSAI (CD-ROM)
|
Journal Code:X0580B
|
ISSN:
|
|
VOL.19th;NO.;PAGE.1B3-02(2005)
|
| Figure&Table&Reference;FIG.5, REF.2 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;In this paper, we propose a map translation method within the framework of the language game. Heterogeneous robots can not use the same map, because they are equipped different sensors and can not recognize the same geographical features or objects. A map translation method is necessary for heterogeneous robots to share spatial knowledge with each other. We assume that robots have probabilistic topological maps. We implemented our methods on a heterogeneous robot simulation. It is shown that two heterogeneous robots learn probabilistic topological maps and the one can estimate the position of the other. This means they can translate maps and are prepared to share spatial knowledge. (author abst.) |
|
|
|
Related Articles;
|
