000181943 001__ 181943
000181943 005__ 20190316235512.0
000181943 037__ $$aCONF
000181943 245__ $$aComputational Aspects of Distributed Optimization in Model Predictive Control
000181943 269__ $$a2012
000181943 260__ $$c2012
000181943 336__ $$aConference Papers
000181943 520__ $$aThis paper presents a systematic computational study on the performance of distributed optimization in model predictive control (MPC). We consider networks of dynamically coupled systems, which are subject to input and state con- straints. The resulting MPC problem is structured according to the system’s dynamics, which makes the problem suitable for distributed optimization. The influence of fundamental aspects of distributed dynamic systems on the performance of two particular distributed optimization methods is systematically analyzed. The methods considered are dual decomposition based on fast gradient updates (DDFG) and the alternating direction method of multipliers (ADMM), while the aspects analyzed are coupling strength, stability, initial state, coupling topology and network size. The methods are found to be sensi- tive to coupling strength and stability, but relatively insensitive to initial state and topology. Moreover, they scale well with the number of subsystems in the network.
000181943 700__ $$aConte, Christian
000181943 700__ $$aSummers, Tyler
000181943 700__ $$0(EPFLAUTH)214806$$g214806$$aZeilinger, Melanie Nicole
000181943 700__ $$aMorari, Manfred
000181943 700__ $$aJones, Colin$$g207237$$0246471
000181943 7112_ $$dDecember 10-13, 2012$$cHawaii, USA$$aConference on Decision and Control
000181943 773__ $$tProceedings of the 51st Conference on Decision and Control
000181943 8564_ $$uhttps://infoscience.epfl.ch/record/181943/files/CDC2012.pdf$$zn/a$$s279497$$yn/a
000181943 909C0 $$0252053$$pLA
000181943 909CO $$pSTI$$ooai:infoscience.tind.io:181943$$qGLOBAL_SET$$pconf
000181943 917Z8 $$x207237
000181943 937__ $$aEPFL-CONF-181943
000181943 973__ $$rREVIEWED$$sACCEPTED$$aEPFL
000181943 980__ $$aCONF