熊本大学情報電気工学科・システム制御研究室

岡島研究室

岡島寛

1980年富山県高岡市出身.博士(工学)大阪大学大学院工学研究科

熊本大学先端科学研究部 准教授

(自然科学教育部情報電気工学専攻,工学部情報電気工学科

論文数 :79 First Author論文数48 平均著者数: 3.03 平均著者順位: 1.67(data

国際会議proceedings 65本

制御工学チャンネルの運営(YouTube, 動画ポータル, ブログ

Hiroshi Okajima (Associate Professor), Kumamoto University Japan

He was born in 1980. He received his M.D. and Ph.D. degrees from Osaka University, Japan, in 2004 and 2007, respectively. He is presently an associate professor at Kumamoto University, Japan. His research interests include control theory.  IEEE CSS, SICE, ISCIE

Web Page (English, Include Research Topics and Achievements)

岡島寛(写真)

共同研究・学術コンサルティング制度・寄付金(支援よろしくお願いします)

システム制御工学研究室との共同研究窓口:熊本大学の共同研究・受託研究・学術コンサルティング制度

私の研究における他にない強みは,制御工学や凸最適化(LMI最適化)の様々な制御分野で成果を出していることが挙げられます.ロバスト制御,ビークルなどの非線形システム制御,マルチレート制御システム,量子化制御,データ駆動制御,状態推定(ソフトウェアセンサ),システム同定(時系列モデル推定)など様々な研究成果があります.特定の制御手法や理論に固執することなく,柔軟なアイディアの提供や研究が可能です.研究業績(論文)はこちら()()をご覧ください.また,学術コンサルティング制度を利用した技術相談や講演も歓迎します.共同研究並びに学術コンサルティングのコミット量や打ち合わせ手段などについて,上記の窓口もしくは岡島までメールを通して気軽にご相談頂ければと思います.

システム制御工学研究室への寄付金:寄附金について | 熊本大学 (kumamoto-u.ac.jp) 

研究費が安定しないと学術研究を継続的に進めることができません.シミュレーションベースで進める研究であっても,計算機環境などの物品費や出張費,論文掲載費(APC)で年間200万円程の研究予算が必要になります(逆に,実機設備を必要としない分,1000万円まであっても使い道がない状況).研究費の多くが年度を跨げないのに対して,寄付金は自由度が高く複数年で計画的に利用することができます.そのため皆様方におかれましては,何卒,ご支援のほどどうぞよろしくお願い致します.支援はこちら(寄付は5000円から可能です.岡島研への支援の場合は,その旨を備考欄に記載下さい.)↓

熊本大学工学部・自然科学科学教育部教育研究支援基金:使途を特定する寄附 - 寄附の種類 | 熊本大学基金 (kumamoto-u.ac.jp) 

また,当研究室に限らず,熊本大学の工学部学生や大学院生に向けた支援(例えば,ロボコンなどの工学部公認サークルの支援等も含む)もこちらの窓口から可能です.どうぞよろしくお願いします.

研究室配属(希望者はぜひ読んで下さい)

岡島研は制御工学の研究室です.制御工学は,自動運転やAI技術などと馴染みの深い研究分野です.当研究室でも,これまで自動運転に関係する研究テーマはいくつか実施しています.以下の研究室紹介動画を見て頂けると研究テーマの雰囲気がわかるかと思います.学生の発表も含めた研究成果一覧はこちらからご覧ください.総合研究棟1201-03が学部生・大学院生の居室になります.研究室の雰囲気を知りたい学部生の方は気軽にお訪ね下さい.

システムインテグレーション研究室(松永研)とキックオフミーティング(4月)と年末報告会(12月)を合同で行っています.制御のものづくりに興味がある方は松永研のHPも見てもらえればと思います.

ミーティングの様子

部屋の様子(研究ミーティング時, 2024.4)

外部ページ

Hiroshi Okajima - リサーチマップ

Hiroshi Okajima (researchgate.net)

H Okajima [制御工学チャンネル] - YouTube (登録者8000名,制御工学)

Hiroshi Okajima - ORCID

Hiroshi Okajima - Google scholar

Hiroshi Okajima - Github(MATLAB code)

システム制御研究室の研究紹介動画(2時間,動画配信システム内でトピック選択可)

熊本大学 松永研究室 (System Integration Lab)

主要な研究論文 Main Research Articles about Control Theory

Hiroshi Okajima, Kenta Arinaga and Atsuo Hayashida, Design of observer-based feedback controller for multi-rate systems with various sampling periods using cyclic reformulation, IEEE ACCESS (2023) (Open Access) 

MATLAB code

Outline: Signal sensing periods typically vary depending on the sensor used and may differ even within a single control system that involves multiple sensors. Likewise, input periods can vary based on the actuator used. This paper discusses the design of observer-based feedback controllers for linear, time-invariant, discrete-time systems operating in a multi-rate sensing and actuating environment. The observation and control periods of the sensors and actuators in the plant are assumed to have mutually rational ratios. First, we reduce the multi-rate system to a periodically time-varying system and provide a linear matrix inequality (LMI) condition for analyzing the l 2 performance using cyclic reformulation, which is a type of time-invariant reformulation for periodic systems. Next, we extend the analysis method to design an observer-based feedback controller for the multi-rate system. This allows us to obtain multi-rate observer gains and feedback gains based on the l 2 -induced norm from disturbances to outputs. Finally, we present numerical results to demonstrate the effectiveness of the observer-based feedback system in the multi-rate environment.

Hiroshi Okajima, Yohei Hosoe and Tomomichi Hagiwara, State Observer under Multi-rate Sensing Environment and Its Design using l2 -Induced Norm, IEEE ACCESS (2023) (Open Access) 

Outline: The duration of a signal's sensing period typically relies on the sensor being used, and can vary even within a single control system that uses multiple sensors. This paper explores the design challenge of creating state observers for linear, time-invariant, discrete-time systems in a multi-rate sensing environment. We assume that the sensing periods of the plant's sensors have mutually rational ratios. First, we describe a state observer for a system with multi-rate sensing as a periodically time-varying state observer. Next, we examine the l2 performance analysis of state estimation errors using the given periodically changing state observer. A linear matrix inequality (LMI) condition is provided for this analysis. By expanding the LMI condition for analysis, we also offer a condition for multi-rate observer synthesis. We then demonstrate the effectiveness of our proposed multi-rate state observer through numerical examples. Notably, even when all sensors share the same period, sensing timing is not unique. As a result, we numerically investigate whether performance varies when the observation timing differs among multiple sensors. [MATLAB code]

Hiroshi Okajima, Yasuaki Kaneda and Nobutomo Matsunaga, State estimation method using median of multiple candidates for observation signals including outliers, SICE Journal of Control, Measurement, and System Integration, Volume 14 Issue 1 pages 257-267 (T&F, Open Access) 

Outline: This paper addresses the state estimation problem for systems with observation outputs that include outliers. The presence of outliers in observation outputs can significantly reduce the accuracy of state estimation. To tackle this issue, we propose a novel observer structure that employs multiple estimated state candidates. First, we generate multiple estimated state candidates, each using the sensing output value from a different detection timing. If outliers are rare, removing candidates affected by outliers helps maintain estimation accuracy. Our proposed observer then selects one estimated state candidate from the group using either a median or weighted median operation. This process ensures that the selected estimated state does not rely on outlier values. Furthermore, we present a method for designing observer gains for these estimated state candidates based on the reachable set of the estimated state error, using Lyapunov-based inequalities. The effectiveness of our proposed observer is demonstrated through numerical examples. [MATLAB Code]

H. Okajima, K. Fujinami, Estimation of Robust Invariant Set for Switched Linear Systems using Recursive State Updating and Robust Invariant Ellipsoid, SICE Journal of Control, Measurement, and System Integration, Volume 14 Issue 1 Pages 97-106 (2021) (T&F, Open Access) 

Outline: This paper presents an analysis method for robust invariant sets in discrete-time linear switched systems with peak-bounded disturbances. Analyzing the robust invariant set in switched linear systems is more challenging than in linear time-invariant systems. We introduce a novel approach to estimate a robust invariant set by combining recursive state updating and an invariant ellipsoid for a common Lyapunov function. The accuracy and effectiveness of our proposed method are demonstrated through numerical examples. [YouTube and  SupportPage] 

Ryuichiro Yoshida, Hiroshi Okajima and Takumi Sato, Model error compensator design for continuous- and discrete-time non-minimum phase systems with polytopic-type uncertainties, SICE Journal of Control, Measurement, and System Integration, Volume 15 Issue 2 pags 141-153 (2022) (T&F, Open Access) 

Outline: This paper introduces a design for a model error compensator combined with a parallel feedforward compensator for continuous- and discrete-time non-minimum phase multiple input multiple output (MIMO) plants. The model error compensator can readily enhance robustness in various control systems. By adding the compensator to the actual plant, the plant's output trajectory can closely follow that of the control system with the intended nominal model. Our previous research proposed a model error compensator design using particle swarm optimization and linear matrix inequalities based on the common Lyapunov function, addressing polytopic-type uncertainties in plants. However, designing the appropriate gain for the model error compensator becomes challenging when dealing with non-minimum phase MIMO systems. In this study, we attach a parallel feedforward compensator to the model error compensator to achieve minimum phase characteristics. With some assumptions, an evaluation system incorporating a parallel feedforward compensator can be derived as a system with polytopic uncertainties. This approach simplifies the gain design of the model error compensator, ensuring robust performance. We demonstrate the effectiveness of our proposed design through numerical examples. [MECpage] 

H. Okajima, K. Sawada and N. Matsunaga:Dynamic Quantizer Design Under Communication Rate Constraints, IEEE Transactions on Automatic Control, Vol.61, No.10, pp.3190-3196 (2016) 

Outline: Feedback-type dynamic quantizers, such as delta-sigma modulators, are often effective for converting high-resolution data into lower-resolution data. These dynamic quantizers consist of a filter and a static quantizer. When controlling under a communication rate constraint, the data rate of the quantizer output should be minimized appropriately through quantization. This technical note presents numerical methods for the comprehensive design of a type of dynamic quantizer, including the selection of all quantizer parameters to minimize a specific performance index while satisfying a communication constraint. We propose a design method for the dynamic quantizer using a particle swarm optimization (PSO) approach. Some initial quantizers in PSO are designed based on an invariant set analysis and an iteration algorithm. The effectiveness of the system employing the proposed quantizer is demonstrated through numerical examples. [SupportPage]

H. Okajima, H. Umei, N. Matsunaga and T. Asai:A Design Method of Compensator to Minimize Model Error, SICE Journal of Control, Measurement, and System Integration, Vol.6, No.4, pp.267-275 (2013) (T&F, Open Access) 

Outline: Robust control design methods have been widely studied in recent decades. A control system performs well under modeling errors and disturbances when the controller design is based on robust control methods. However, it is well known that control systems typically face a trade-off between control performance and robustness. To address this trade-off problem, an internal model-type compensator structure that minimizes the modeling gap between the nominal model and actual plant dynamics is proposed.   By employing the proposed compensator, the dynamics of the compensated system closely match those of the nominal model. Additionally, we introduce a design method for compensator parameters aimed at minimizing a set of plant dynamics. The proposed design method can be reduced to the standard µ design control problem. Using the proposed compensator for control systems instead of the plant itself may lead to improved output performance despite plant uncertainty. As the proposed compensator can be used for controlling not only linear but also nonlinear plants, it allows for the easy achievement of robust control in nonlinear systems. The effectiveness of our proposed method is demonstrated through numerical examples. [MECpage]  

H. Okajima and T. Asai:Performance Limitation of Tracking Control Problem for a Class of References, IEEE Transactions on Automatic Control, Vol.56, No.11, pp.2723-2727 (2011) 

Outline: This technical note focuses on the analysis of fundamental limitations in tracking control problems for single-input single-output (SISO) systems. Existing results analyze these limitations based on specific assumptions of reference signals, such as step, trigonometric signals, and others. In contrast, we define a class of reference signals in a more abstract but general way. For this general class of reference inputs, we provide an analytical solution for tracking performance limitations based on the achievable set of outputs, which are characterized by transfer functions. The analysis results clearly separate the contributions of the plant and the reference signal. [SupportPage and PDF]

岡島寛:状態のロバスト不変集合を用いた外れ値除去機能を有するMCVオブザーバの解析とオブザーバゲイン設計,計測自動制御学会論文集,Vol. 55, No. 12, pp. 823-829 (2019) 

岡島寛,細江陽平,萩原朋道: マルチレート系の状態推定のための周期時変状態オブザーバのl2誘導ノルム評価による設計,計測自動制御学会,Vol. 55, No. 12, pp. 792-799 (2019) 

岡島寛:有限整定条件下での不安定零点と最小アンダーシュート量の解析的表現,計測自動制御学会論文集,Vol. 55, No. 10, pp. 648-654 (2019) 

岡島寛,川脇治:制御システムを個性化するための制御系設計,システム制御情報学会論文誌,Vol. 31, No. 11, pp. 392-399 (2018)

岡島寛,西村悠樹,松永信智:モデル誤差抑制補償に基づく非線形システムのフィードバック線形化,計測自動制御学会論文集,Vol.50, No.12, pp.869-874 (2014) j-stage

鍋倉司樹,岡島寛,松永信智:プレフィルタとポストフィルタを含むAD/DA変換系の設計と音声信号圧縮系への適用,計測自動制御学会論文集,Vol.50, No.3, pp.295-302 (2014) j-stage

岡島寛,松永信智,澤田賢治:通信容量制約に基づく動的量子化器の量子化幅設計と性能解析,計測自動制御学会論文集,Vol.46,No.6,pp.327-335 (2010) j-stage 

岡島寛,松永信智,川路茂保:ニュートラルステア特性実現のための直接ヨーモーメント制御,計測自動制御学会論文集,Vol.45,No.3,pp.153-159 (2009) j-stage 

岡島寛,浅井徹,川路茂保:離散時間系に対する最適追従制御問題の解析解,計測自動制御学会論文集,Vol.44,No.10, pp.793-801 (2008) j-stage 

岡島寛,浅井徹,川路茂保:経路追従問題における最適速度制御,計測自動制御学会論文集,Vol.44,No.7,pp.566-574 (2008) j-stage 

岡島寛,浅井徹:「軌道の差」の評価に基づく軌道追従制御,システム制御情報学会論文誌,Vol.20,No.4,pp.133-143 (2007) j-stage 

モデル誤差抑制補償器

Model Error Compensator

動的量子化器

Dynamic Quantizer

マルチレートシステム

Multi-rate state observer