ブールネットワークによるロボット制御の概念実証
Boolean network robotics: a proof of concept
ブールネットワークをロボットの行動制御に用い、確率的局所探索で自動設計したネットワークが光刺激による空間移動と内部記憶の形成・利用を可能にすることを示した。
著者: Andrea Roli, Mattia Manfroni, Carlo Pinciroli, Mauro Birattari
分類: cs.AI, cs.NE, cs.RO
原文アブストラクト
Dynamical systems theory and complexity science provide powerful tools for analysing artificial agents and robots. Furthermore, they have been recently proposed also as a source of design principles and guidelines. Boolean networks are a prominent example of complex dynamical systems and they have been shown to effectively capture important phenomena in gene regulation. From an engineering perspective, these models are very compelling, because they can exhibit rich and complex behaviours, in spite of the compactness of their description. In this paper, we propose the use of Boolean networks for controlling robots' behaviour. The network is designed by means of an automatic procedure based on stochastic local search techniques. We show that this approach makes it possible to design a network which enables the robot to accomplish a task that requires the capability of navigating the space using a light stimulus, as well as the formation and use of an internal memory.
