Kousuke INOUELecturer

■Researcher basic information

Organization

  • College of Engineering Department of Mechanical Systems Engineering
  • Graduate School of Science and Engineering（Master's Program） Major in Mechanical Systems Engineering
  • Faculty of Applied Science and Engineering Domain of Mechanical Systems Engineering

Research Areas

  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering), Control and systems engineering, System Engineering
  • Informatics, Robotics and intelligent systems, Intelligent Mechanics and Machine System
  • Informatics, Intelligent informatics, Intelligent Informatics

Research Keyword

  • bio-mimetic robotics, multi-robot cooperation, robot learning

Degree

  • 2002年03月　博士（工学）（東京大学）
  • 1998年03月　修士（工学）（東京大学）

Educational Background

  • 2002, The University of Tokyo, Graduate School, Division of Engineering, Department of Precision Machinery Engineering

Career

  • Apr. 2010, 茨城大学, 工学部 知能システム工学科, 講師
  • Apr. 2007, Research Associate: Dept. of Intelligent Systems Engineering, Ibaraki University
  • May 2002 - Mar. 2007, 茨城大学 工学部 システム工学科・助手
  • Apr. 2002 - Apr. 2002, 東京大学 人工物工学研究センター 研究期間研究員

External link

Message from Researchers

  • (Message from Researchers)

    (研究経歴)
    1996/04-1998/03 東京大学 工学部 精密機械工学科 修士課程学生として，複数移動ロボットの協調的繰り返し搬送作業の制御に関する研究に従事
    1998/04-2002/03 東京大学 大学院 工学系研究科 精密機械工学専攻 博士課程学生として，部分観測環境における移動ロボットの行動学習の研究に従事
    2002/04 東京大学 人工物工学研究センター 研究機関研究員として，サービス工学の研究に従事
    2002/05- 茨城大学 工学部 システム工学科において，複数移動ロボットの協調制御，ロボット行動学習，生物模倣型ロボットの研究に従事

■Research activity information

Award

  • Nov. 2000, 平成12 年度 IMS (Intelligent Manufacturing Systems) 論文賞, Jun OTA, Tamio ARAI, Kousuke INOUE, Ryousuke CHIBA, Tomokazu HIRANO: Flexible Transport System by Cooperation of Conveyer-Loaded AGVs, Proceedings of 2000 IEEE International Conference on Robotics and Automation (ICRA2000), pp.1144-1150, (2000/10)

Paper

  • Behavior Acquisition in Partically Observable Environments by Autonomous Segmentation of the Observation Space
    Kousuke INOUE; Tamio ARAI; Jun OTA, In this paper, we propose a method by which an agent can autonomously construct a state-representation to achieve state-identification with a sufficient Markovian property. Furthermore, the agent does this using continuous and multi-dimensional observationspace in partially observable environments. In order to deal with the non-Markovian property of the environment, a state-representation of a decision tree structure based on past observations and actions is used. This representation is gradually segmented to achieve appropriate state-distinction. Because the observation-space of the agent is not segmented in advance, the agent has to determine the cause of its state-representation insufficiency: (1) insufficient observation-space segmentation, or (2) perceptual aliasing. In the proposed method, the cause is determined using a statistical analysis of past experiences, and the method of state-segmentation is decided based on this cause. Results of simulations in twodimensional grid-environments and experiments with real mobile robot navigating in two-dimensional continuous workspace shows that an agent can successfully acquire navigation behaviors with many hidden states., 日本機械学会
    Journal of Robotics and Mechatronics, Jun. 2015, [Reviewed]
  • Development of information collection system with omnidirectional images
    Naoji Shiroma; Tomohiro Kai; Kousuke Inoue; Yasuhiro Fukuoka; Yoshikazu Mori; Kenji Itou
    2015 INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATICS AND BIOMEDICAL SCIENCES (ICIIBMS), 2015, [Reviewed]
  • 現存する脊椎動物の神経系を搭載した恐竜ロボットによる2足動歩行の実現
    赤間淳貴; 福岡泰宏; 森善一; 城間直司; 井上康介; 中野博民, In this paper, we try to achieve biomimetic biped walking of a dinosaur-like robot, focusing on its nervous system as well as its mechanism. We have been developing a dinosaur robot "Raptetsu" based on the studies of dinosaurs on archaeology and also apply typical biomimetic locomotion control methods used for existing vertebrates for the robot. In detail, we utilize a CPG (Central Pattern Generator), which is a locomotion rhythm generator that exists in vertebrates' spinal cord, to generate the robot's walking rhythm. Moreover, reflex centers is set below the CPG in the spinal cord and produce joint torque of two legs in the swing and stance phases. Thus, we hope to produce an adaptive dynamic walking generated by interaction between the original mechanism of dinosaurs and existing animals' nervous system and try to contribute to the understanding of dinosaurs' locomotion on archaeology by observing and analyzing the dynamic walking experimental results., The Japan Society of Mechanical Engineers
    日本機械学会論文集 (Ｃ編), Oct. 2011, [Reviewed]
  • Acceleration of Reinforcement Learning by a Mobile Robot using Generalized Rules　　　　　　　　　　　　　　　
    Kousuke INOUE; Jun OTA; Tomohiko KATAYAMA; Tamio ARAI, One very fundamental problem in behavioral learning by an agent is that it takes quite a long time to acquire optimal behavior. In order to solve this problem, in this paper, we propose an approach to make learning processes more efficient by the use of generalized knowledge. In this approach, the agent repeats learning processes for different tasks and extracts behavioral rules that are commonly harmful to task execution by the use of statistical method. After sufficient experience is accumulated, the generalized rules are extracted from the experience and are applied to subsequent learning processes, and, consequently, the learning processes are accelerated by inhibiting commonly harmful behaviors. In order to achieve generality of rule expression, the description of the rules is based on egocentric information, namely, raw data of observations and actions experienced by the agent. In order to avoid a perceptual aliasing problem, the rule expression includes information on sequential experience and a mechanism is introduced to control the balance of utility and generality of the rules. The proposedmethod is examined in navigation tasks by amobile robot in grid environments as an example of application. The results show that the proposed method accelerates learning processes.
    Proceedings of 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2000), Apr. 2010, [Reviewed]
  • 車体屈折式操行車両の非線形直線経路追従制御
    城間 直司; 石川 哲史; 井上 康介; 福岡 泰宏; 森 善一, Autonomous control of an articulated steering type vehicle such as a wheel loader, which is usually used at surface mining fields to load mineral resources and rocks, is highly expected for operation cost reduction. A wheel loader is one of heavy machines which is also used for snow removing and/or loading operation at construction fields because of its high mobility. Its steering system is articulated steering and the rear wheels follow tracks where the front wheels have traveled. We have introduced a new virtual velocity constraint of the vehicle and formulated nonlinear state equations using two velocity constraints. A nonlinear state-feedback controller is designed using exact linearization. Simulation and experimental results show that an articulated steering type vehicle can follow a straight line with the proposed feedback control method., The Japan Society of Mechanical Engineers
    日本機械学会論文集 (C編), Mar. 2010, [Reviewed]
  • Acceleration of Reinforcement Learning by a Mobile Robot Using Generalized Inhibition Rules　　　　　　　　　　　　　　　
    Kousuke INOUE; Jun OTA; Tamio ARAI, One very fundamental problem in behavioral learning by an agent is that it takes quite a long time to acquire optimal behavior. In order to solve this problem, in this paper, we propose an approach to make learning processes more efficient by the use of generalized knowledge. In this approach, the agent repeats learning processes for different tasks and extracts behavioral rules that are commonly harmful to task execution by the use of statistical method. After sufficient experience is accumulated, the generalized rules are extracted from the experience and are applied to subsequent learning processes, and, consequently, the learning processes are accelerated by inhibiting commonly harmful behaviors. In order to achieve generality of rule expression, the description of the rules is based on egocentric information, namely, raw data of observations and actions experienced by the agent. In order to avoid a perceptual aliasing problem, the rule expression includes information on sequential experience and a mechanism is introduced to control the balance of utility and generality of the rules. The proposedmethod is examined in navigation tasks by amobile robot in grid environments as an example of application. The results show that the proposed method accelerates learning processes.
    Journal of Robotics and Mechatronics, Feb. 2010, [Reviewed]
  • Iterative Transportation by Multiple Mobile Robots Considering Unknown Obstacles　　　　　　　　　　　　　　　
    Kousuke INOUE; Jun OTA; Tamio ARAI
    Journal of Robotics and Mechatronics, Feb. 2009
  • Autonomous Behavior Generator for a Companion Robot: SELF　　　　　　　　　　　　　　　
    Yoshikazu MORI; Naoyuki KUBOTA; Kousuke INOUE
    Kansei Engineering International, Jan. 2009, [Reviewed]
  • A study of a merging section of PRT system with bulk arrival and its control strategy
    Takahiro Hoshino; Kousuke Inoue; Kazuhiro Tsuboi; Yoshio Hamamatsu
    IEEJ Transactions on Industry Applications, 2008, [Reviewed]
  • ****　　　　　　　　　　　　　　　
    Takahiro HOSHINO; Kousuke INOUE; Kazuhiro TSUBOI; Yoshio HAMAMATSU
    電気学会論文誌, 01 Jan. 2008, [Reviewed]
  • Development of a Novel Crawler Mechanism with Polymorphic Locomotion
    Gualangping LAN; Shugen MA; Kousuke INOUE; Yoshio HAMAMATSU
    Advanced Robotics, Feb. 2007, [Reviewed]
  • A ROBUST SENSOR FOR DOWNY MILDEW DISEASE BASED ON IMAGE PROCESSING　　　　　　　　　　　　　　　
    Anshukha SRIVASTAVA; Masatake SHIRAISHI; Kousuke INOUE
    New Agriculturist, 2007, [Reviewed]
  • Influence of Gradient of a Slope to Optimal Locomotion Curves of a Snake-like Robot
    Shugen MA; Naoki TADOKORO; Kousuke INOUE
    Advanced Robotics, Feb. 2006, [Reviewed]
  • Several insights into omnidirectional static walking of a quadruped robot on a slope
    Lei Zhang; Shugen Ma; Kousuke Inoue
    2006 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-12, 2006, [Reviewed]
  • 3次元蛇型ロボットの動力学解析　　　　　　　　　　　　　　　
    大豆生田; 吉弘; 馬 書根; 井上 康介
    日本機械学会論文集C編, Apr. 2004, [Reviewed]
  • Behavior Acquisition for Multiple Tasks in a Partially Observable Environment
    Kousuke INOUE; Jun OTA; Tamio ARAI, This paper proposes a method for acquisition of behavior capable of multiple tasks by an agent with non-predesigned and continuous observation space in partially observable environment. In the proposed method, the agent discriminates the ongoing task from other tasks using experience acquired in acquisition process of behaviors corresponding to the tasks, and applies knowledge for the corresponding task. In order to cope with lack of clues for task-identi.cation and bad in.uence on the task-execution behavior by the task-identifying behavior, additional learning process is executed as the need arises. Simulation results shows the proposed method can build a suitable behavior for multiple tasks in a simple partially observable grid-environment., The Society of Instrument and Control Engineers
    Trans. SICE, Jul. 2002, [Reviewed]
  • コンベア搭載型AGVの協調による物体搬送システム　　　　　　　　　　　　　　　
    太田 順; 新井 民夫; 井上 康介; 千葉 龍介; 平野 智一; 前田 雄介
    日本機械学会論文集C編, Jun. 2001, [Reviewed]
  • Material Transport System by Cooperation of Conveyor-Loaded AGVs
    Jun Ota; Tamio Arai; Kousuke Inoue; Ryousuke Chiba; Tomokazu Hirano; Yusuke Maeda
    Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C, 2001, [Reviewed]
  • 視覚情報を用いた状態・行動空間の自律的生成
    小林 祐一; 太田 順; 井上 康介; 新井 民夫, 計測自動制御学会
    計測自動制御学会論文集, Nov. 2000, [Reviewed]
  • 未知環境における移動ロボット群の経路学習　　　　　　　　　　　　　　　
    平野 智一; 太田 順; 井上 康介; 倉林 大輔; 新井 民夫
    日本機械学会論文集C編, Feb. 2000, [Reviewed]
  • Path Learning by Robot Group in Unknown Environment
    Tomokazu Hirano; Jun Ota; Kousuke Inoue; Daisuke Kurabayashi; Tamio Arai
    Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C, 2000, [Reviewed]
  • 群ロボットによる多数物体の繰り返し搬送計画
    吉村 裕司; 太田 順; 井上 康介; 平野 智一; 倉林 大輔; 新井 民夫, This paper focuses on a planning method for an iterative transportation task by cooperative mobile robots. This task requires the generation of appropriate robot paths and the formation of groups of cooperating robots. In order to realize efficient transportation, the planning architecture consisting of "Path-Generation Phase" and "Strategy-Making Phase" is proposed. The former phase generates robot paths from global environmental information and produces a graph network from the derived robot paths. In the latter phase, every robot learns a behavior strategy based on the derived graph. An asymptotic strategy-making method is used here. The global transportation strategy derived by the proposed architecture can be divided into three types, and simulation results indicate that forming the global transportation type realizes efficient transportation. Experimental results indicate that the proposed architecture is practical enough for real situation., The Robotics Society of Japan
    日本ロボット学会誌, May 1998, [Reviewed]

MISC

Books and other publications

  • Insect Mimetics Handbook　　　　　　　　　　　　　　　
    Eds: Tateo SHIMODATE; Takahiko HARIYAMA, Joint translation
    NTS Press, 2009
  • 知の創成－身体性認知科学への招待－ (原著: R. Pfeifer, C. Scheier: Understanding Intelligence, MIT Press, (1999))　　　　　　　　　　　　　　　
    R. Pfeifer; C. Sheier著; 石黒 章夫; 小林 宏; 細田 耕 監訳, Joint translation
    共立出版, Nov. 2001

Lectures, oral presentations, etc.

  • ヘビ身体のエッジ構造が運動に及ぼす影響に関する構成論的調査,　　　　　　　　　　　　　　　
    立花 佳大,井上 康介
    第34回自律分散システム・シンポジウム, Jan. 2022, 計測自動制御学会
    202201, 202201
  • ヘビのエッジ機構の構成論的調査　　　　　　　　　　　　　　　
    立花 佳大; 井上 康介
    第29回茨城講演会, Aug. 2021, 日本機械学会
    202108, 202108
  • ヘビのペグ押し推進を再現するロボットの開発 ~自律分散制御システムの構築とペグ押し推進の解析~　　　　　　　　　　　　　　　
    林 忠文; 井上 康介
    第29回茨城講演会, Aug. 2021, 日本機械学会
    202108, 202108
  • ヘビの隘路における運動メカニズムの調査　　　　　　　　　　　　　　　
    大嶋 冬偉; 井上 康介
    第29回茨城講演会, Aug. 2021, 日本機械学会
    202108, 202108
  • トレッドミルモデルに基づくヤスデの歩容混在メカニズムの構成論的調査　　　　　　　　　　　　　　　
    寺川 京佑; 賀澤 柊弥; 豊田 晋久; 井上 康介
    第33回自律分散システム・シンポジウム, Jan. 2021, 計測自動制御学会
    202101, 202101
  • 多足類模倣型ロボットの開発と適応的な歩行の実現　　　　　　　　　　　　　　　
    豊田 晋久,井上 康介
    第33回自律分散システム・シンポジウム, Jan. 2021, 計測自動制御学会
    202101, 202101
  • ヘビの側方くねり運動と押しつけ運動間遷移の再現　　　　　　　　　　　　　　　
    村越 洋介; 井上 康介
    第32回自律分散システム・シンポジウム, Jan. 2020, 計測自動制御学会
    202001, 202001
  • シミュレーションによるヤスデの歩行制御系の調査　　　　　　　　　　　　　　　
    渡邉 季誠; 井上 康介
    ロボティクス・メカトロニクス 講演会 2019, Jun. 2019, 日本機械学会
    201906, 201906
  • 連続的な腹部機構を有するヘビ型ロボットの開発　　　　　　　　　　　　　　　
    大木 寿馬; 井上 康介
    ロボティクス・メカトロニクス 講演会 2019, Jun. 2019, 日本機械学会
    201906, 201906
  • 歩行ロボットの安定歩行のための新しい足部機構の開発　　　　　　　　　　　　　　　
    許 方,井上 康介,森 善一
    第26回茨城講演会, Aug. 2018, 日本機械学会
    201808, 201808
  • ヘビの筋骨格系が側方くねり運動に与える影響の調査　　　　　　　　　　　　　　　
    稲葉 雄哉,井上 康介,桐林 颯,森 善一
    第26回茨城講演会, Aug. 2018, 日本機械学会
    201808, 201808
  • エンジニアがやる生物学 ─ 長い動物の自律分散制御メカニズムの解明に向けて ─　　　　　　　　　　　　　　　
    井上康介
    SICE中部支部 制御理論研究委員会・組込みシステムと制御研究委員会 合同研究会, 17 Dec. 2014, 計測自動制御学会 (SICE) 中部支部 制御理論研究委員会, [Invited]
  • エンジニアがやる生物学─ 長い動物の自律分散制御メカニズムの解明に向けて ─　　　　　　　　　　　　　　　
    井上康介
    SICE中部支部 制御理論研究委員会・組込みシステムと制御研究委員会 合同研究会, 17 Dec. 2014, 計測自動制御学会 (SICE) 中部支部 制御理論研究委員会, [Invited]
  • 連続的かつ多次元の観測空間を持つエージェントによる部分観測環境における自律的状態空間構成　　　　　　　　　　　　　　　
    井上 康介; 太田 順; 新井 民夫
    第19回日本ロボット学会学術講演会, Sep. 2001, 日本ロボット学会
    200109, 200109
  • 局所センサ入力に基づく移動ロボットのナビゲーション行動の学習　　　　　　　　　　　　　　　
    井上 康介; 太田 順; 小林 祐一; 湯浅 秀男; 新井 民夫
    第13回自律分散システム・シンポジウム, Jan. 2001, 計測自動制御学会
    200101, 200101
  • 局所情報の利用に基づく強化学習による移動ロボットのナビゲーションの学習　　　　　　　　　　　　　　　
    井上 康介; 太田 順; 湯浅 秀男; 新井 民夫
    第18回日本ロボット学会学術講演会, Sep. 2000, 日本ロボット学会
    200009, 200009
  • 汎化ルールによる強化学習の加速　　　　　　　　　　　　　　　
    井上 康介; 片山 朋彦; 太田 順; 新井 民夫
    ロボティクス・メカトロニクス講演会'00, May 2000, 日本機械学会
    200005, 200005
  • 部分観測環境下における自律的状態分割による強化学習　　　　　　　　　　　　　　　
    井上 康介; 太田 順; 小林 祐一; 新井 民夫
    第12回自律分散システム・シンポジウム, Jan. 2000, 計測自動制御学会
    200001, 200001
  • 部分観測環境下における強化学習による移動ロボットの行動獲得　　　　　　　　　　　　　　　
    井上 康介; 太田 順; 千葉 龍介; 新井 民夫
    第11回自律分散システム・シンポジウム, Jan. 1999, 計測自動制御学会
    199901, 199901
  • 群ロボットによる多数物体の搬送計画（第5報: 未知環境における繰り返し搬送計画）　　　　　　　　　　　　　　　
    井上 康介; 太田 順; 平野 智一; 倉林 大輔; 新井 民夫
    第15回日本ロボット学会学術講演会, Sep. 1997, 日本ロボット学会
    199709, 199709

Affiliated academic society

  • Jan. 2002, 計測自動制御学会
  • Apr. 1997, 日本ロボット学会