Open AccessModel-Based Control of a BCF Mode Carangiform Bioinspired Robotic Fish
Author(s) -
Abhra Roy Chowdhury,
Vinoth Kumar,
Bhuneshwar Prasad,
Rajesh Kumar,
Sanjib Kumar Panda
Publication year - 2014
Publication title -
marine technology society journal/marine technology society journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.23
H-Index - 43
eISSN - 1948-1209
pISSN - 0025-3324
DOI - 10.4031/mtsj.48.4.5
Subject(s) - fish locomotion , control theory (sociology) , modular design , dynamic positioning , thrust , matlab , control engineering , fish fin , engineering , actuator , simulation , controller (irrigation) , torque , computer science , marine engineering , kinematics , fish <actinopterygii> , control (management) , artificial intelligence , mechanical engineering , fishery , biology , operating system , agronomy , physics , thermodynamics , classical mechanics
Bioinspired robotic locomotion in the ocean environment can unveil critical issues on maneuverability, efficiency, and power consumption. This paper describes the modeling and closed-loop control of a bioinspired robotic fish. A body-caudal fin (BCF) carangiform swimming mode is presented. The propulsion scheme simulates the oscillatory motion of fish tail as thrust generator. The manufactured prototype is a 45-cm-long BCF mode four-joint, 6 degree of freedom modular robotic fish with a horizontal caudal fin (tail). The system uses DC servomotors as actuators and is controlled by microcontroller dsPIC33F. The mechanical CAD design in done in Solidworks and its 3D motion simulations in Matlab VRML, respectively. Lagrange-based dynamic modeling is done for the robotic fish. Based on the model, two nonlinear closed-loop control schemes, namely computed torque method and feed-forward control, both with dynamic PD compensation, are evaluated. This paper compares these model-based controllers to match the desired response based on reference angle position and velocity tracking. Real-time simulation results in Matlab/Simulink are provided to illustrate the effectiveness of the proposed methodologies for robotic fish locomotion.