WEKO3
アイテム
Merging Navigation Strategies for Vehicles by Structured Boltzmann Machine
http://hdl.handle.net/10445/6203
http://hdl.handle.net/10445/6203b0a15622-97a8-42f6-9432-308635b99fde
| Item type | 会議発表論文 / Conference Paper(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2011-04-19 | |||||
| タイトル | ||||||
| タイトル | Merging Navigation Strategies for Vehicles by Structured Boltzmann Machine | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_5794 | |||||
| 資源タイプ | conference paper | |||||
| アクセス権 | ||||||
| アクセス権 | metadata only access | |||||
| アクセス権URI | http://purl.org/coar/access_right/c_14cb | |||||
| 著者 |
奥野, 拓
× 奥野, 拓× Kakazu, Yukinori |
|||||
| 抄録 | ||||||
| 内容記述タイプ | Abstract | |||||
| 内容記述 | This paper deals with the decision making of a mobile robot in drastic changes of its envi-ronment. For the problem, we are trying to incorporate the rule decomposi-tion-and-merging approach into the rein-forcement learning approach. This idea is realized modifying the framework of the Structured Boltzmann Machine (SBM). The SBM represents structured concepts and re-trieves stored concepts by complex keys. For the latter the SBM augments operations on the network modules connecting the independently learned modules and extracting concepts by specified keys. Such operations are adapted and applied to merge strategies learned in advance. A simplified vehicle navigation problem will be defined and a modified SBM specific to that problem will be illustrated. | |||||
| 書誌情報 |
Proceedings of the IEEE Intelligent Vehicles '93 Symposium p. 355-359, 発行日 1993-10 |
|||||
| 査読有無 | ||||||
| 値 | あり/yes | |||||
| 研究業績種別 | ||||||
| 値 | 国際会議/International Conference | |||||
| 単著共著 | ||||||
| 値 | 共著/joint | |||||
