Motoyasu TANAKA (田中 基康)

Offer Organization: 日本学術振興会, System Name: 科学研究費助成事業, Category: 基盤研究(B), Fund Type: -, Overall Grant Amount: - (direct: 9900000, indirect: 2970000)
（１）柔軟環境のモデル化：①紐状の柔軟物をモデル化し，ロボットが摩擦力を効果的に発生させることで紐を登るタスクを検討対象とした．紐状柔軟物とロボットとの摩擦力は，体幹に巻き付く紐の巻き角と力により推定できる．②柔軟環境として粉体環境（砂地など）を想定し，生物のヘビの運動を参考にしながら，潜る動作を一つモデル化した．
（２）移動剛体環境での環境適応制御：実センサの特性を考慮し，前年度提案した制御手法に改良を施した．固定柔軟環境での環境適応制御：断熱材上のフィルムの偏りを軽減する動作手法を提案し，シミュレーションで検証を行った．
（３）近接覚センサの完成：（２）の検証で用いるための近接覚センサのキャリブレーションを実施し，実機搭載が可能な状態にした．触覚センサの開発：ヘビ型ロボットが柔軟環境との接触状態を認識するための触覚センサの開発を行った．巻き角と力を推定可能である．検出方式には気圧式を採用し，32個の気圧センサ素子を2段の放射状に配置し，ウレタンゴムとシリコーンゴムを用いた被膜方法によって，紐状柔軟物との巻き付きに必要な外形と摩擦特性をもち，その全域の接触を検知可能な構成を実現した．成果は次年度に学会発表予定である．
（４）検証実験：近接覚センサを搭載した実機を開発し，前述（２）の検証実験を実施し，提案制御則の有効性を確認した．
（計画外）①粉体環境について実機実験を基にした，粉体への潜行動作最適化のための方法について考察，実験のための環境構築と予備的な実験，簡略化された物理モデルに基づき，潜行動作の物理的妥当性について議論するための検討を行った．②ヘビ型ロボットの全身と環境との相対位置を認識するためのシステムおよびアルゴリズムの検討を開始した．SLAMの性能を損なうことなく，センサの実装位置の不確かさに同時に対処可能なアルゴリズムを開発した．

Offer Organization: 日本学術振興会, System Name: 科学研究費助成事業 基盤研究(B), Category: 基盤研究(B), Fund Type: -, Overall Grant Amount: - (direct: 9900000, indirect: 2970000)
(1)可変環境を分類し，モデル化の検討を開始：可変環境のうち，複数の移動可能な剛体が存在する環境を考え，剛体の移動や利用についての検討を行った．ある一定以上の力を加えると動きだす剛体と接触しているヘビ型ロボットの運動を考え，その運動方程式を導出したほか，所望の運動を達成するための制御則について考察を行った．
(2)移動剛体環境における環境適応制御の検討を開始：未知の２平面環境を対象に，近接覚センサを用いた環境適応制御手法を提案した．センサ情報をもとに制御が行われるため，ロボットの位置姿勢を外部センサによって検出する必要がない．また，移動中にロボットと環境との適切な接触を保つため，センサ情報をもとに復帰動作を行うことができる．成果は次年度に学会発表予定である．
(3)環境を紐に限定した移動柔軟環境における環境改変制御を提案：紐の昇降制御を提案した．紐の形状を巧みに利用し滑落を回避できる．提案手法は論文投稿中である．
(4)環境変化を読み取るための近接覚センサの開発を開始：前述(2)に適する近接覚センサとして，ヘビ型ロボットの形状を著しく変更せずに実装でき，ロボットの近傍にある平面の情報を取得可能なセンサを開発した．未知環境での使用を想定しているため，対象の材質の影響を受けにくいものとした．さらに，検出範囲を簡単に調整可能であるため，使用環境，および方法に適切な検出範囲の検討を短期間で実施することができる．
(5)前述(3)の検証実験：ヘビ型ロボットを開発して紐の昇降制御の検証実験を行い，提案制御則の有効性を確認した．検証結果は論文投稿中である．

Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C), Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3400000, indirect: 1020000)
This report deals with modeling and control of a small unmanned aircraft. First, a nonlinear dynamical model is constructed from the equations of motion of a considered small unmanned aircraft and the aerodynamic characteristics generated by its wings. Furthermore, we develop a new design framework to stabilize the small unmanned aircraft. In addition, a rational polynomial fuzzy controller, which is an extension of the polynomial fuzzy controller, is also proposed, and control system design methods using these controllers are proposed. After verifying the effectiveness of the proposed design methods through simulations, flight experiments are conducted using the actual small unmanned aerial vehicle.

Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C), Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3400000, indirect: 1020000)
We proposed a hyper-redundant control method for the cord-like mechanical system which can bend and stretch. In the method, the change of the constrained condition with the surrounding terrain is represented as the change of the system matrices, and we can use a redundancy caused by bending and stretching of the robot. The effectiveness of the method was verified by simulations and experiments. As for the consideration of the locomotion intelligence of a living snake, we discussed the ascending and descending motions of a snake-like robot while clinging to the surface of a cylinder, using its undulating motion.

Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C), Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3500000, indirect: 1050000)
This reports mainly presents three new results with respect to nonlinear control for a flying robot that is a kind of fixed-wing type unmanned aerial vehicles. The first result is a nonlinear model construction including the aerodynamics and actuators dynamics. The second result is a guaranteed cost control design for the constructed nonlinear model. The controller is designed by minimizing the upper bound of a given quadratic performance function. Hence, the design can be regarded as a nonlinear optimal control design. The third result is that the utility of our modeling and control is demonstrated through real flight experiments.

Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B), Category: Grant-in-Aid for Scientific Research (B), Fund Type: -, Overall Grant Amount: - (direct: 13500000, indirect: 4050000)
This study addresses the development of a robot for inspection of old bridges. By suspending the robot with a wire and controlling the wire length, the movement of the robot is realized. The robot mounts a high-definition camera and aims to detect cracks on the concrete surface of the bridge using this camera. An inspection method using an unmanned aerial vehicle (UAV) has been proposed. Compared to the method using an unmanned aerial vehicle, the wire-suspended robot system has the advantage of insensitivity to wind and ability to carry heavy equipment. This makes it possible to install a high-definition camera and a cleaning function to find cracks that are difficult to detect due to dirt.

Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research Grant-in-Aid for Young Scientists (B), Category: Grant-in-Aid for Young Scientists (B), Fund Type: -, Overall Grant Amount: - (direct: 3000000, indirect: 900000)
In this study, to propose a control method moving on uneven terrain for a snake robot, we consider following four main topics. (1) Obstacle avoidance of whole body of a snake robot through a multiple obstacle environment. (2) Trajectory tracking on curved surface and multiple planes. (3) Terrain following method by using the robot's own weight on uneven terrain. (4) As application (1)-(3), cleaning control of an articulated cleaning robot on a narrow path and a step and stair climbing control of an articulated mobile robot.