On-line learning to recover from thruster failures on autonomous underwater vehicles

File Description SizeFormat 
Leonetti_OCEANS-2013.pdfAccepted version525.41 kBAdobe PDFView/Open
Title: On-line learning to recover from thruster failures on autonomous underwater vehicles
Authors: Leonetti, M
Ahmadzadeh, SR
Kormushev, P
Item Type: Conference Paper
Abstract: We propose a method for computing on-line the controller of an Autonomous Underwater Vehicle under thruster failures. The method is general and can be applied to both redundant and under-actuated AUVs, as it does not rely on the modification of the thruster control matrix. We define an optimization problem on a specific class of functions, in order to compute the optimal control law that achieves the target without using the faulty thruster. The method is framed within model-based policy search for reinforcement learning, and we study its applicability on the model of the AUV Girona500. We performed experiments with policies of increasing complexity, testing the on-line feasibility of the approach as the optimization problem becomes more complex.
Issue Date: 30-Sep-2013
Date of Acceptance: 23-Sep-2013
URI: http://hdl.handle.net/10044/1/26074
Publisher: IEEE
Start Page: 1
End Page: 6
Journal / Book Title: Proc. MTS/IEEE Intl Conf. OCEANS 2013
Copyright Statement: © 2013 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Conference Name: OCEANS 2013
Publication Status: Published
Publisher URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6741265
Start Date: 2013-09-23
Finish Date: 2013-09-27
Conference Place: San Diego, CA
Appears in Collections:Dyson School of Design Engineering



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx