成果報告書詳細
管理番号20160000000742
タイトル*平成27年度中間年報 次世代ロボット中核技術開発 革新的ロボット要素技術分野 生物ロコモーションの本質理解から切り拓く大自由度ロボットの革新的自律分散制御技術
公開日2016/9/10
報告書年度2015 - 2015
委託先名国立大学法人東北大学
プロジェクト番号P15009
部署名ロボット・AI部
和文要約
英文要約Title: Innovative Decentralized Control Technology for Large Degree-of-Freedom Robot Inspired by Animal Locomotion (FY2015-FY2016) FY2015 Annual Report
Most robots are designed to work in predefined environments, and uncertainties such as terrain irregularities often reduce their functionalities. Engineers have treated these uncertainties as disturbances and have coped with them by implementing complicated control algorithms. In contrast, animals do not lose their functionality under unstructured and unpredictable real-world constraints, and they adapt to unforeseen circumstance in real time by appropriately self-organizing their behavior. The clarification of this remarkable mechanism will bridge the gap between robots and animals and will help develop robots that work well in undefined environments.
Snakes are a suitable model for investigating these mechanisms. For snakes, irregularities are in fact beneficial for their locomotion: snakes actively utilize terrain irregularities and produce an effective propulsion force by pushing their body against the scaffolds that they encounter (hereafter, we refer to this locomotion as “scaffold-based locomotion”). However, the mechanism underlying scaffold-based locomotion still remains elusive.In this project aiming at developing highly adaptive robots that could not be realized solely by conventional methods, we first tried to model the control mechanism underlying scaffold-based locomotion of snakes. We proposed an autonomous decentralized control scheme on the basis of a concept called “TEGOTAE,” a Japanese concept describing how well a perceived reaction, i.e., sensory information, matches an expectation, i.e., intention of controller. In order to validate the TEGOTAE-based decentralized control rule, we built a modular snake-like robot consisting of 36 body segments, and then succeeded in reproducing scaffold-based locomotion in the real-world test environments.
In the following year, we will further investigate the performance of the proposed TEGOTAE-based decentralized control mechanism quantitatively in terms of adaptability, fault tolerance, scalability, and operability. We will also investigate decentralized control schemes that allow our robots to exhibit various types of locomotion, e.g., concertina, sidewinding, in response to the environment that they encounter.
ダウンロード成果報告書データベース(ユーザ登録必須)から、ダウンロードしてください。

▲トップに戻る