Energy-Shaping Gait Generation for a Class of Underactuated Robotic Fish

Daniel Roper*, Sanjay Sharma, Robert Sutton, Philip Culverhouse

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

<B>Abstract</B><P></P>The recent explosion of interest in subsea research and industries has led to a demand for efficient, versatile submarine robots to perform interaction and survey tasks. While most of these robots employ conventional propeller- or thruster-based locomotion, there has also been a growing interest in the development of biologically inspired robotic swimmers (or robotic fish). Such devices take inspiration from biological swimmers in the hope of gaining some of their manoeuvrability and efficiency.<P></P>Early robotic fish achieved their gait by directly controlling the relative angle between each vertebrae using multiple active actuators.<P></P>Following observations that many biological locomotive gaits utilize spring dynamics to create efficient oscillatory motion with minimal active input, there has been a recent trend toward the development of underactuated robotics utilizing dynamics to achieve harmonic locomotive gaits.<P></P>For a harmonic dynamic system, the path is dependent on the total energy in the system. By controlling the total energy in the system, the gait can then be controlled by varying a single degree of freedom.<P></P>This study explores the potential of a novel approach to use an energy-based control to produce an effective propulsive gait for a class of robotic fish. It outlines a deadbeat control strategy and presents simulation results, which demonstrate the effectiveness of the approach. Furthermore, it is shown that the gaits produced would unlikely be found using geometric gait optimization.
Original languageEnglish
Pages (from-to)34-43-34-43
Number of pages0
JournalMarine Technology Society Journal
Volume46
Issue number0
DOIs
Publication statusPublished - 1 May 2012

Fingerprint

Dive into the research topics of 'Energy-Shaping Gait Generation for a Class of Underactuated Robotic Fish'. Together they form a unique fingerprint.

Cite this