Biomimetic Swimming Analysis and Motion Control of Muti-joint Robotic Fish

• Main Authors: Li-Yuan Yeh, 葉禮源
• Other Authors: Leehter Yao
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/7autww

碩士 === 國立臺北科技大學 === 自動化科技研究所 === 106 === In order to make the robotic fish achieve a highly biomimetic swimming and accurate underwater motion control, this thesis firstly collects the real swimming data of the real salmon, performs bionic analysis, and then designs and plans a comprehensive hardware and algorithm for the robotic fish. In the hardware part of the practice, this article actually measures the shape of real fish, and according to its streamlined structure and the softness of the tail fin design a bionic mechanical architecture, so that the robotic fish in the water with the least resistance, produce a great propulsion. Biomimetic analysis is performed through the video of squid swimming to perform image processing to collect the changes in joint angle during swimming. The Fourier series is used for data fitting (curve fitting) to obtain the equation of motion of each action. Applying it to the swimming of robotic fish, it can realize the bionic nature underwater and swim the highly efficient movement posture. Finally, in order to enable the robotic fish to achieve different degrees of motion control, this paper strengthens the analysis of the joint data of fish movements, and establishes the exercise parameters, and finally realizes on the robotic fish, so that the robotic fish can reach different levels of underwater bionic movement results. The experimental results show that the robot fish has achieved excellent results in linear swimming efficiency, swimming stability, and bionic motion control.