Introduction to Distributed Control
Prev: Distributed Estimation | Course Home | Next: Estimation and Control in Networked Systems |
In this lecture, we take a look at the problem of distributed control. We will begin by seeing why the problem is hard. Then we will look at one obvious approach towards solving the problem. Other approaches to the problem will also be mentioned.
Lecture Materials
Reading
Most of the cited papers are available using IEEE Xplore[1]. If you are unable to obtain any, please send me [2] a mail.
Reference 2 in the lecture notes covers the concept of information patterns.
"Separation of Estimation and Control for Discrete Time Systems", H. S. Witsenhausen, Proceedings of the IEEE, vol. 59, no. 11, pp. 1557-1566, Nov. 1971.
The synthesis algorithm we will cover in class is described in this paper.
"On the Synthesis of Control Laws for a Network of Autonomous Agents", V. Gupta, B. Hassibi and R. M. Murray, Proceedings of the American Control Conference 2004, vol. 6, pp. 4927-4932, 2004.
Additional synthesis methods can be found in the references. I particularly recommend references 14, 21 and 22.
- "Distributed Control Design for Systems Interconnected over an Arbitrary Graph", C. Langbort, R. S. Chandra and R. D'Andrea, IEEE Transactions on Automatic Control, vol. 49, no. 9, pp. 1502-1519, Sep. 2004.
- "Optimal Control of Systems with Delayed Observation Sharing Patterns Via Input-Output Methods", P. G. Voulgaris, Systems and Controls Letters, Vol. 50, pp. 51-64, 2003.
- "A Characterization of Convex Problems in Decentralized Control", M. Rotkowitz and S. Lall, IEEE Transactions on Automatic Control, vol. 51, no. 2, pp.274-286, Feb. 2006.