User Tools

Site Tools


dynamics_and_control_of_distributed_systems

Differences

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

Link to this comparison view

dynamics_and_control_of_distributed_systems [2016/09/01 19:15] (current)
Line 1: Line 1:
 +Title: Dynamics and control of distributed systems
  
 +Speaker: Mihailo Jovanovic, University of Minnesota
 +
 +Abstract:
 +
 +In the first part of the talk, we examine fundamental limitations
 +arising from the use of local feedback in networks subject to
 +stochastic disturbances. For vehicular formation control problems in
 +topology of regular lattices we show that it is impossible to have
 +coherent large formations, that behave like rigid lattices, in one and
 +two spatial dimensions. Yet we prove that this is achievable in 3D.
 +The observed phenomenon is a consequence of the fact that, in 1D and
 +2D, local feedback laws are ineffective in guarding against
 +disturbances with large spatial wavelength. We provide connections
 +with several other problems including distributed averaging
 +algorithms, global mean first passage time of random walks, effective
 +resistance in electrical networks, and statistical mechanics of
 +harmonic solids.
 +
 +In the second part of the talk, we demonstrate how tools and ideas
 +from control theory, optimization,​ and compressive sensing can be
 +combined to identify network topologies that strike desired tradeoff
 +between the performance and sparsity. Our approach consists of two
 +steps. First, we identify sparsity patterns of the feedback gains by
 +incorporating sparsity-promoting penalty functions into the optimal
 +control problem, where the added terms penalize the number of
 +communication links in the distributed controller. Second, we optimize
 +feedback gains subject to structural constraints determined by the
 +identified sparsity patterns. In the first step, the sparsity
 +structure of feedback gains is identified using the alternating
 +direction method of multipliers,​ an algorithm well-suited to large
 +optimization problems. Several examples are provided to demonstrate
 +the effectiveness of the developed approach.
 +
 +Bio sketch:
 +
 +Mihailo Jovanovic (www.umn.edu/​~mihailo) is an Associate Professor of
 +Electrical and Computer Engineering at the University of Minnesota,
 +Minneapolis,​ where he also serves as the Director of Graduate Studies
 +in the interdisciplinary PhD program in Control Science and Dynamical
 +Systems. He has held visiting positions with Stanford University and
 +the Institute for Mathematics and its Applications. His current
 +research focuses on sparsity-promoting optimal control, fundamental
 +performance limitations in the design of large dynamic networks, and
 +dynamics and control of fluid flows. He is a senior member of IEEE,
 +and a member of APS and SIAM. He currently serves as an Associate
 +Editor of the SIAM Journal on Control and Optimization and has served
 +as an Associate Editor of the IEEE Control Systems Society Conference
 +Editorial Board from July 2006 until December 2010. He received a
 +CAREER Award from the National Science Foundation in 2007, an Early
 +Career Award from the University of Minnesota Initiative for Renewable
 +Energy and the Environment in 2010, a Resident Fellowship within the
 +Institute on the Environment at the University of Minnesota in 2012,
 +and the George S. Axelby Outstanding Paper Award from the IEEE Control
 +Systems Society in 2013.
 +----------------------------------------------------------------------
dynamics_and_control_of_distributed_systems.txt ยท Last modified: 2016/09/01 19:15 (external edit)