User Tools

Site Tools



This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
vidi:track2 [Monday, 16 March 2009 : 11:48:59]
vidi:track2 [Monday, 16 March 2009 : 12:33:59] (current)
Line 1: Line 1:
 +======Research proposal======
 +=====Plan of work=====
 +====Research plan====
 +===Track 2: Observer design for networks and distributed control===
 +The idea of deriving observers for switching networks and using those for
 +deriving distributed controllers still leaves many questions that need to be
 +answered. Roughly the following four stages can be distinguished.
 +First, observability needs to be properly defined and tests for observability
 +have to be developed. In the above mentioned example both servers are able to reconstruct the state at the other
 +server based on only local information. However, in an acyclic network, information only
 +flows downstream. Therefore, only the "​final"​ server in the network is able to reconstruct
 +the global state information. The "​first"​ server in the network can not reconstruct anything
 +from the global state. A good starting point might be to use a behavioral
 +approach for dynamical systems as presented in  ([#​References|references]).
 +Second, observers need to be designed for arbitrary observable networks.
 +As a small deviation in the measurements
 +due to noise should also lead to a small observer error,
 +a general methodology needs to be developed to arrive at observers which
 +reconstruct the global state, but are not too sensitive to noise.
 +Related to this is also the stability analysis of distributed policies, since
 +the servers are not necessarily synchronized,​ which is the next issue to address.
 +This will be joint research with prof. Armbruster
 +(Arizona State University, Tempe, AZ; part-time with Systems Engineering Group, Mechanical Engineering,​ TU/e),
 +including applications to self-organizing networks.
 +For that purpose, the PhD Student should also stay at Arizona State University, Tempe, AZ, for three months,
 +around the end of his/her second year.
 +Finally, distributed controllers should be designed for networks, which is not simply a combination
 +of state feedback controllers and observers.
 +Notice that even though for an acyclic network the "​first"​ server can not
 +reconstruct anything, the "​second"​ server is able to reconstruct the state at the "​first"​
 +server and therefore might be able to adjust itself properly to the "​first"​ server.
 +Eventually, the entire network might still converge towards desired network behavior.
vidi/track2.txt · Last modified: Monday, 16 March 2009 : 12:33:59 by lefeber