Model predictive and non-cooperative dynamic game fault recovery control strategies for a network of unmanned underwater vehicles
- 10 August 2017
- journal article
- research article
- Published by Taylor & Francis in International Journal of Control
- Vol. 92 (3) , 489-517
Abstract
In this paper, the problem of control and fault recovery for a team of autonomous underwater vehicles in the presence of loss of effectiveness (LOE) actuator faults is addressed. Towards this end, two different fault recovery control strategies based on the model predictive control technique as well as the dynamic game theory are proposed and developed. Given the allowable information that can be exchanged among the agents, both centralised and semi-decentralised recovery control schemes are considered and their associated corresponding fault recovery strategies are developed. The proposed active fault recovery control strategies incorporate both the online inaccurate as well as delayed actuator fault estimates to reconfigure the nominal (healthy state) controllers. The effectiveness of the proposed semi-decentralised fault recovery control schemes is quantitatively investigated through extensive simulation case studies considering various LOE actuator fault severities in one or more unmanned vehicles as well as fault detection and isolation module imperfections such as fault estimation error and time delays in detecting the faults. The simulation results demonstrate and illustrate that our proposed semi-decentralised recovery control scheme can maintain acceptable degraded tracking and formation keeping performance of both the faulty and healthy agents in the team with lower computational and communication bandwidth requirements as well as lower or fairly close control effort cost as compared to the centralised control recovery scheme.Keywords
Funding Information
- Qatar Foundation (5-045-2-017)
This publication has 33 references indexed in Scilit:
- Cooperative actuator fault accommodation in formation flight of unmanned vehicles using absolute measurementsIET Control Theory & Applications, 2012
- Multi-agent team cooperation: A game theory approachAutomatica, 2009
- Fault Detection and Isolation of discrete-time Markovian jump linear systems with application to a network of multi-agent systems having imperfect communication channelsAutomatica, 2009
- Bibliographical review on reconfigurable fault-tolerant control systemsAnnual Reviews in Control, 2008
- Distributed Predictive Control: Synthesis, Stability and Feasibility2007
- Decentralized receding horizon control for large scale dynamically decoupled systemsAutomatica, 2006
- A sub-optimal algorithm to synthesize control laws for a network of dynamic agentsInternational Journal of Control, 2005
- Adaptive control of an autonomous underwater vehicle: experimental results on ODINIEEE Transactions on Control Systems Technology, 2001
- Position and attitude tracking of AUV's: a quaternion feedback approachIEEE Journal of Oceanic Engineering, 1994
- Linear model predictive control of unstable processesJournal of Process Control, 1993