Abstract:An exploitation of prior knowledge in parameter estimation becomes vital whenever measured data is not informative enough. Elicitation of quantified prior knowledge is a well-elaborated art in societal and medical applications but not in the engineering ones. Frequently required involvement of a facilitator is mostly unrealistic due to either facilitator's high costs or complexity of modelled relationships that cannot be grasped by humans. This paper provides a facilitator-free approach based on an advanced kn… Show more
“…It is presented in [31]. Let us note that it was proposed by the first author of [15] and successfully applied in [32]. Loosely, it follows from an application of minimum cross-entropy principle [33], [34] and its generalisation [35] allowing non-linear constraints on pds to be optimised according to this principle.…”
Section: B Sharing Of Knowledge Brought By Predictorsmentioning
confidence: 99%
“…A S is given by (32) and where κ t+1 is a Gaussian noise with covariance matrix 0.001I, and U t+1 = [u 1,t+1 , u 1,t+1 ] is the vector of control inputs. The parameters of the lattice given in (34) are assumed to be unknown and recursively estimated.…”
Abstract-Computational and communication complexities call for distributed, robust, and adaptive control. This paper proposes a promising way of bottom up design of distributed control in which simple controllers are responsible for individual nodes. The overall behaviour of the network can be achieved by interconnecting such controlled loops in cascade control for example, and by enabling the individual nodes to share information about data with their neighbours without aiming at unattainable global solution. The problem is addressed by employing a fully probabilistic design, which can cope with inherent uncertainties, that can be implemented adaptively and which provide a systematic rich way to information sharing. The paper elaborates the overall solution, applies it to linear-Gaussian case and provides simulation results.
“…It is presented in [31]. Let us note that it was proposed by the first author of [15] and successfully applied in [32]. Loosely, it follows from an application of minimum cross-entropy principle [33], [34] and its generalisation [35] allowing non-linear constraints on pds to be optimised according to this principle.…”
Section: B Sharing Of Knowledge Brought By Predictorsmentioning
confidence: 99%
“…A S is given by (32) and where κ t+1 is a Gaussian noise with covariance matrix 0.001I, and U t+1 = [u 1,t+1 , u 1,t+1 ] is the vector of control inputs. The parameters of the lattice given in (34) are assumed to be unknown and recursively estimated.…”
Abstract-Computational and communication complexities call for distributed, robust, and adaptive control. This paper proposes a promising way of bottom up design of distributed control in which simple controllers are responsible for individual nodes. The overall behaviour of the network can be achieved by interconnecting such controlled loops in cascade control for example, and by enabling the individual nodes to share information about data with their neighbours without aiming at unattainable global solution. The problem is addressed by employing a fully probabilistic design, which can cope with inherent uncertainties, that can be implemented adaptively and which provide a systematic rich way to information sharing. The paper elaborates the overall solution, applies it to linear-Gaussian case and provides simulation results.
“…• Currently available FPD procedures for a) merging of external knowledge [20,26], b) approximate recursive learning and stabilized forgetting [16,17,18], c) decision strategy design [2,36], and d) local adaptive control design [24], are unified for the first time via the hierarchical FPD framework of this paper.…”
Section: Accepted Manuscriptmentioning
confidence: 99%
“…Adopting M = M o , as asserted in (20), which is equivalent to substituting S(A|K) = S o (A|K) (19) in the right-hand side of ( 22), then…”
“…Specifically, porous macromolecular materials have attracted durable attention of the scientific community [ 58 , 63 , 64 , 65 , 66 ] due to their functionality and possibility of mechanical control. Nevertheless, this is quite challenging, for a general task to predict and keep under control [ 67 , 68 , 69 ] the desired properties of the self-assembling materials. Several attempts for investigation of stretched porous polyethylene (PE) filled with LC compounds have been already done [ 56 , 70 , 71 , 72 , 73 ].…”
The control and prediction of soft systems exhibiting self-organization behavior can be realized by different means but still remains a highlighted task. Novel advanced nanocomposite system has been designed by filling of a stretched porous polyethylene (PE) film with pore dimensions of hundreds of nanometers by chiral ferroelectric liquid crystalline (LC) compound possessing polar self-assembling behavior. Lactic acid derivative exhibiting the paraelectric orthogonal smectic A* and the ferroelectric tilted smectic C* phases over a broad temperature range is used as a self-assembling compound. The morphology of nanocomposite film has been checked by Atomic Force Microscopy (AFM). The designed nanocomposite has been studied by polarizing optical microscopy (POM), differential scanning calorimetry (DSC), small and wide-angle X-ray scattering and broadband dielectric spectroscopy. The effect of a porous PE confinement on self-assembling, structural, and dielectric behavior of the chiral LC compound has been established and discussed. While the mesomorphic and structural properties of the nanocomposite are found not to be much influenced in comparison to that of a pure LC compound, the polar properties have been toughly suppressed by the specific confinement. Nevertheless, the electro-optic switching was clearly observed under applied electric field of low frequency (210 V, 19 Hz). The dielectric spectroscopy and X-ray results reveal that the helical structure of the ferroelectric liquid crystal inside the PE matrix is completely unwound, and the molecules are aligned along stretching direction. Obtained results demonstrate possibilities of using stretched porous polyolefins as promising matrices for the design of new nanocomposites.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.