Bio-harmonized control experiments of a carangiform robotic fish underwater vehicle. (3rd March 2016)

• Record Type: Journal Article
• Authors: Chowdhury, Abhra Roy; Sasidhar, Sangit; Panda, S. K.
• Is Part Of: Advanced robotics. Volume 30:Number 5(2016)
• Page Start: 338
• Page End: 351
• Publication Date: 2016-03-03
• Subjects: Bio-inspired underwater robotics -- carangiform -- lighthill slender body theory -- dynamics modeling -- non-linear control -- computed-torque method -- feed-forward method.; Robotics -- Periodicals; Robotics -- Japan -- Periodicals; Robotics; Japan; Periodicals; 629.89205
• DOI: 10.1080/01691864.2015.1114905
• Languages: English
• ISSNs: 0169-1864

Abstract

Abstract : This paper presents experimental implementation and comparison of three different control schemes of a bio-inspired robotic fish underwater vehicle. The dynamics model is obtained by unifying conventional rigid body dynamics and bio-fluid dynamics of a carangiform fish swimming given by Lighthill's(LH) slender body theory. It proposes an inclusive mathematical design for better control and energy efficient path travel for the robotic fish. The system is modeled as an two-link robot manipulator (caudal tail) with a mobile base (head). This forward thrust drives the robotic fish head represented by a combined non-linear equation of motion in earth fixed frame. We develop and compare the dynamic motion closed loop control strategy of the bio-harmonized robotic fish based on three different non-linear control schemes using CTM (Computed Torque Method), FF (Feed-Forward) controllers both with dynamic PD compensation and finally a proposed combination of CTM with FF. An inverse dynamic control method based on non-linear state function model including hydrodynamics is proposed to improve tracking performance. CTM control generates a feedback loop for linearization and decoupling robot dynamic model with a shorter response time, while a dynamic PD compensation in the FF path is employed by FF scheme through the desired trajectories. FF model-based strategy results in an improved tracking and shorter route to travel between two points. Overall results indicate that performances of the proposed control schemes based on the inverse dynamic model are comparable and useful for robotic fish motion tracking in fluid environment. Graphical Abstract: