RBFネットワークを利用した超小型人工衛星ドッキング姿勢時におけるオンライン画像推定法(機械力学,計測,自動制御)

概要

論文の詳細を見る
Recently, much attention has been focused on application of teaching and learning algorithm such as radial basis function (RBF) networks. Since it is expected that visual recognition problem, especially expressional and obstacle recognition, the estimated results can be cope with their situated environments without human associations as possible. On the other hand, The conventional navigating system for satellite in space has been suffered from the problem of huge check sequence we needed as the complexity of task increases. In this paper, availability of on-line learning with RBF network is discussed to apply for the image recognition of the posuture in which micro satellite (I.e daughter ship) is approaching to the docking port of the mother ship. The ability for all-purpose of RBF is useful for uncertain inputs and the method for the image estimation in docking phase is established experimentally under the 6 DOF environment.
2004-10-25