Development of Computational Intelligence Aided Control Method -Toward Application to Large Scale Complex System in IoT Era-

2019 Fiscal Year Final Research Report

• Project/Area Number: 16K06409
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Control engineering/System engineering
• Research Institution: University of Tsukuba
• Principal Investigator: KAWABE Tohru 筑波大学, システム情報系, 教授 (40224844)
• Project Period (FY): 2016-04-01 – 2020-03-31
• Keywords: 計算知能援用制御 / IoT / 大規模複雑システム / モデル予測制御 / 機械学習 / V2G / パーソナルモビリティ

Outline of Final Research Achievements

As a basis of a new control system design method with a correction function based on computational intelligence mainly for deep learning, we extended the control method combining PID control and sliding mode control with model predictive control. A linear multiple regression model constructed from the results of analyzing a large amount of video data and a control method based on the inverse problem of optimal control based on this model are also developed. Furthermore, a motion control method for an autonomous vehicle that extends the potential function into space and time and calculates the predicted trajectory of obstacles based on data and learning is developed. The practicality of these development methods was verified for large-scale complex systems such as smart grids and next-generation urban transportation systems in pedestrian coexistence spaces.

Free Research Field

システム制御工学

Academic Significance and Societal Importance of the Research Achievements

ビッグデータを学習し、その解析結果を活かした制御法を構築するための基盤理論を、モデル予測制御を各種制御法と組み合わせ、従来の制御法の枠組みを超えて拡張することで実用性を高めた点に学術的意義がある。また、再生可能エネルギー源と電気自動車を結んだ電力送電網の運用制御手法を開発したことにより、CO2排出量を削減し安定かつ耐久性に優れた次世代の電力網の実現や、次世代都市交通システムの一つであるパーソナルモビリティのための運動制御法を開発したことにより、新たな大規模交通システムへの実用化が期待できることに社会的意義がある。