2A1-G14 行動知能の漸次構成による複雑なロボット協調行動の実現(複数ロボットの協調制御)

概要

論文の詳細を見る
This work is concerned with realization of complex cooperative robot actions by gradual construction of action intelligence. According to the Intelligent Composite Motion Control (ICMC), complex actions are gradually realized from fundamental motions. The action intelligence of a complex cooperation is constructed in multi-stage manner using the Multi-stage Genetic Algorithm, MGA. The MGA solves the cooperative action realization problem as recursive combinatorial optimization. The empirical knowledge obtained is effectively utilized to solve similar problems efficiently and moreover for more complex actions. The method is successfully applied to optimal realization of cooperative robot soccer actions. For further development, versatile cooperation planning using variable-chromosome-length GA, introduction of robot character, and utilization of tools are also investigated.
一般社団法人日本機械学会の論文
2008-06-06