|
|
Metody spolupráce v bayesovském rozhodování s víceúčastníky - rukopis PhD. práce
Kracík, Jan
This work has its origin in the GAAV project 1ET100750401 BADDYR – Bayesian Adaptive Distributed Decision Making – solved in the Department of Adaptive Systems in the Institute of Information Theory and Automation in years 2004-2007. The objective of the BADDYR project was to develop a theoretical and algorithmic background for a distributed decision making with multiple Bayesian decision makers. The results of the BADDYR project are further developed in the GAˇCR project 102/08/0567 – Fully probabilistic design of dynamic decision strategies. This work contributes to these projects by a design of methods for communication of the decision makers.
|
|
|
O strukturách prediktorů umožňujících distribuované dynamické Bayesovské rozhodování
Šmídl, Václav
Decentralized adaptive control is based on the use of many local controllers in parallel, each of them estimating its own local model and pursuing its local aims. If each controller designs its strategy using only its own model, the resulting control may be poor since consequences of actions of the neighbors are not taken into account. We seek a way how to improve algorithm of decision strategy design of a single local controller without significant increase in complexity of the local model or complexity of the design procedure. In this paper we study variants of distributed dynamic programming that could be evaluated locally. Specifically, we will investigate variants of the fully probabilistic control strategy design. Distributed and cen- tralized control strategies will be compared.
|
|
|
O bayesovském rozhodování s více účastníky
Kracík, Jan
This report briefly summarizes the perspective on decision making with multiple Bayesian decision makers which has been reached during the work on the GAAV project 1ET100750401 BADDYR. It gradually turned out that the problem is much deeper than it may appear. At the same time we foud out that in some special, yet practically important, cases a suitable solution could be reached withou any novel methods just via detailed modelling of knowledge and preferences of individual particiapns. This special cases also serve us as a motivation for solution of more general problems.
|
|
|
Spojování znalostí jako problém odhadování
Kárný, Miroslav ; Bodini, A. ; Ruggeri, F.
Bayesian cooperative decision making by multiple participants requires a combination of environment models of cooperating neighbors. In a sense, a speci¯c participant tries to exploit imprecise probabilistic knowledge con- cerning of low-dimensional marginal and conditional distributions o®ered by his neighbors. In spite of the substantial overlap with domains like knowl- edge elicitation, estimation of graphical models or building of probabilistic expert systems, the merging problem was not addressed yet in its entirety.
|
|
| |
|
|
Matlab-Aimsun Toolbox 2.1
Dohnal, Pavel ; Dibelka, Lukáš ; Elbl, Marek
This text is the user's guide for installing and using toolbox for performing traffic experiments in Matlab 7.1 (or higher). The simulation is automatically ran in extern application - traffic simulator Aimsun 4.2. This is part of software package Getram Extensions 4.2, which must be installed, because is necessary for toolbox function. Getram Extensions 4.2 is supported only for Win32 platform, so that toolbox can be installed only on this platform (Windows 2000, Windows XP, etc.)
|
|
|
Toolbox pro dopravní simulace
Dohnal, Pavel
The Aimsun simulator is powerful tool for simulation of traffic flow while the Matlab environment provides the rich set of functions that simplify the process of design, optimization and testing of developed traffic control. Toolbox combines together advantages of both of them.
|
|
|
Příklady odhadu stavu a parametrů pro lineární model s rovnoměrně rozloženými inovacemi
Pavelková, Lenka
In this contribution, state-space model with uniformly distributed innovations is introduced and the Bayesian state estimation proposed. The off-line evaluation of the maximum a posteriori probability (MAP) estimate of unknowns in the linear state-space model with uniform innovations reduces to linear programming (LP). The solution provides either estimates of the noise boundary and parameters or of the noise boundary and states. The on-line estimation is obtained by applying LP on the sliding window, i.e., by considering only the fixed amount, say partial, of the newest last data and states items. By swapping between state and parameter estimations, joint parameter and state estimation is obtained. The use of Taylor expansion for approximation of products of unknowns solves also the joint parameter and state estimation. Simulation studies help to get an insight on the potential and restrictions of these heuristic method. This contribution shares the experimentally gained experience with both these solutions of the joint state and parameter estimation.
|
|
|
Toolbox Jobcontrol s GUI
Novák, Miroslav ; Tesař, Ludvík
Jobcontrol is a toolbox merging a user-friendly graphical user interface with a powerful set of utilities for system identification employing Gaussian mixture model. It is implemented as a set of M-scripts and MEX-binary executables for the Matlab computing environment. It suits to the goal of finding a suitable structure for the given data and then it creates the optimal controller by tuning its parameters to fit the user's requirements.
|
|
|
Výpočetní aspekty návrhu regulátoru a vyčíslení kvality
Novák, Miroslav
This work contributes to the activity in the Department of Adaptive Systems in Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic to develop a complete design algorithm for advanced controllers such as the LQG one and put them through to real applications. The task of controller tuning is to transform the user specified requirements into the values of the tuning parameters. The system knowledge is incomplete. The Bayesian estimation delivers the parameters not as known numbers but as their probability density function. The important contribution of this work is extending the tuning to the multiple input multiple output (MIMO) controllers, where multiple constraints on particular quantities are considered simultaneously.
|
guest ::
