Skip to main content
Thesis defences

PhD Oral Exam - Zahra Gallehdari, Electrical and Computer Engineering

Cooperative Control Reconfiguration in Networked Multi-Agent Systems


Date & time
Thursday, September 29, 2016
9:30 a.m. – 12:30 p.m.
Cost

This event is free

Organization

School of Graduate Studies

Contact

Sharon Carey
514-848-2424, ext. 3802

Where

Engineering, Computer Science and Visual Arts Integrated Complex
1515 St. Catherine W.
Room EV 2.184

Wheel chair accessible

Yes

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Development of a network of autonomous cooperating vehicles has attracted significant attention during the past few years due to its broad range of applications in areas such as (a) autonomous underwater vehicles for exploring deep sea oceans, (b) satellite formations for space missions, and (c) mobile robots in industrial sites where human involvement is impossible or restricted, to name a few. Motivated by the stringent specifications and requirements for depth, speed, position or attitude of the team and the possibility of having unexpected actuators and sensors faults in missions for these vehicles have led to the proposed research in this thesis on cooperative fault-tolerant control design of autonomous networked vehicles.

First, a multi-agent system under a fixed and undirected network topology and subject to actuator faults is studied. A reconfigurable control law is proposed and the so-called distributed Hamilton-Jacobi-Bellman equations for the faulty agents are derived. Then, the reconfigured controller gains are designed by solving these equations subject to the faulty agent dynamics as well as the network structural constraints to ensure that the agents can reach a consensus even in presence of a fault while simultaneously the team performance index is minimized.

Next, a multi-agent network subject to simultaneous as well as subsequent actuator faults and under directed fixed topology and subject to bounded energy disturbances is considered. An H∞ performance fault recovery control strategy is proposed that guarantees (a) the state consensus errors remain bounded, (b) the output of the faulty system behaves exactly the same as that of the healthy system, and (c) the specified H∞ performance bound is guaranteed to be minimized. Towards this end, the reconfigured control law gains are selected first by employing a geometric control approach where a set of controllers guarantees that the output of the faulty agent imitates that of the healthy agent and the consensus achievement objectives are satisfied. Then, the remaining degrees of freedom in the selection of the control law gains are used to minimize the bound on a specified H∞ performance index.

Then, control reconfiguration problem in a team subject to directed switching topology networks as well as actuator faults and their severity estimation uncertainties is considered. The consensus achievement of the faulty network is transformed into two stability problems, in which one can be solved offline while the other should be solved online and by utilizing information that each agent has received from the fault detection and identification module. Using quadratic and convex hull Lyapunov functions the control gains are designed and selected such that the team consensus achievement is guaranteed while the upper bound of the team cost performance index is minimized.

Finally, a team of non-identical agents subject to actuator faults is considered. A distributed output feedback control strategy is proposed which guarantees that agents output follow the outputs of the exo-system and the agents states remains stable even when agents are subject to actuator faults.


Back to top

© Concordia University