1,721,358 research outputs found
"Class-Type" identification-based internal models in multivariable nonlinear output regulation
Full text linkThe paper deals with the problem of output regulation in a “non-equilibrium” context for a special class of multivariable nonlinear systems stabilizable by high-gain feedback. A post-processing internal model design suitable for the multivariable nature of the system, which might have more inputs than regulation errors, is proposed. Uncertainties in the system and exosystem are dealt with by assuming that the ideal steady state input belongs to a certain “class of signals" by which an appropriate model set for the internal model can be derived. The adaptation mechanism for the internal model is then cast as an identification problem and a least square solution is specifically developed. In line with recent developments in the field, the vision that emerges from the paper is that approximate, possibly asymptotic, regulation is the appropriate way of approaching the problem in a multivariable and uncertain context. New insights about the use of identification tools in the design of adaptive internal models are also presented
Privacy-preserving and inconsistency-tolerant query answering in knowledge bases and databases
No full textIn recent decades, the widespread use of information systems has led to an increasing amount of sensitive data managed, posing several privacy and security risks. Controlled Query Evaluation (CQE) is a framework for limiting access to sensitive information when querying a system, while still providing useful responses. Initially proposed for databases, CQE has recently been studied for Description Logics ontologies, the logical formalization underpinning most Semantic Web technologies.
Central to CQE is the possibility of specifying a data protection policy in a declarative way, avoiding the need to actively alter data to ensure their security. The policy is enforced via optimal censors, which aim to minimally alter information while maintaining confidentiality. Notably, censors satisfying the so-called indistinguishability property also ensure that the end user is not even able to discern if the answers to a query have been altered or not.
The thesis investigates various CQE semantics based on the so-called optimal GA censors, which enjoy the indistinguishability property. In order not to arbitrarly pick a censor, we adopt the usual strategy of skeptical reasoning, i.e. answering according to all the optimal censors. Unfortunately, this approach is proven to be intractable, so we present an alternative semantics based on the intersection of all optimal GA censors (called IGA censor). Under this semantics, answering union of conjunctive queries (UCQs) in the Description Logic DL-LiteR is first-order (FO) rewritable, and thus in AC0 in data complexity.
Anyway, the intersection-based approach may cause losing too much information, which led us to explore smarter semantics for improving the throughput of query answers while preserving confidentiality. Towards this aim, we first show how to exploit a priority relation between ontology predicates for reducing the number of censors. Then, we consider the strategy of selecting censors dynamically, exploiting the order in which queries are posed for maximizing the cooperativity of the system. In both cases, we provide suitable FO rewriting techniques, proving that the nice computational properties of the IGA semantics are preserved.
The prioritized scenario also served as main setting for carrying out our experiments, wherein we implemented the newly-presented CQE techniques adapting them to the ontology-based data access methodology.
Moreover, the research extends to the related field of Consistent Query Answering (CQA), which studies how to handle inconsistencies when answering queries. Specifically, we consider knowledge bases in which a set of existential rules must always be satisfied by the underlying database. A central notion in CQA is the one of repair, that is a maximal subset of the database satisfying all the rules. As for query answering, similarly as done in CQE, we examine both skeptical reasoning and the intersection-based semantics, known as AR and IAR, respectively. We study a very expressive language of rules and identify many subclasses where repair checking and UCQ entailment are tractable, or even FO rewritable.
We finally investigate the integration between open and closed-world assumption, offering insights into utilizing open and closed predicates within the CQA framework, and provide the first set of complexity results for the aforenamed decision problems within this intriguing scenario
Necessary Conditions for Output Regulation with Exosystem Modelled by Differential Inclusions
Full text linkThe problem of output regulation has always been tackled in frameworks in which the references to be tracked and the disturbances to be rejected are generated by an autonomous differential equation, referred to as the exosystem. This assumption, that is routinely used in the design of asymptotic regulators, plays also a fundamental role in the formulation of the regulation problem and in the definition of the basic concepts such as the steady state and the zero dynamics of nonlinear systems. In this paper we show that the concepts of steady state, zero dynamics and the output regulation problem can be equivalently defined in a framework in which the exosystem is generated by a differential inclusion
Robust Nonlinear Regulation: Continuous-Time Internal Models and Hybrid Identifiers
No full textThe paper deals with the problem of robust output regulation for minimum-phase nonlinear systems in a semiglobal setting. We present a different perspective to the problem of adaptive regulation in which prediction error identification methods, which are routinely used in other control contexts, can be adopted to design robust nonlinear regulators. The proposed control structure combines continuous-time dynamics and 'hybrid identifiers', the latter specifically designed to estimate the actual steady-state control law. The proposed framework encompasses existing frameworks proposed so far in the nonlinear continuous-time literature
Isolating Invisible Dynamics in the Design of Robust Hybrid Internal Models
No full textThe paper deals with the problem of output regulation for a class of hybrid linear systems and exosystems whose state jumps periodically according to a known clock. In this framework the main contribution of the paper is to present a general method for the systematic design of robust internal model-based regulators by extending results that are known for continuous-time systems. The internal model design procedure relies upon a notion of visibility of the so-called “hybrid steady-state generator.” The general theory is applied to the case of robust tracking of spline-based reference trajectories by showing how the latter can be thought of as being generated by hybrid linear exosystems
Output regulation of nonlinear systems by sliding mode
No full textThis paper focuses on the design of an output feedback sliding mode regulator able to achieve the asymptotic tracking of a reference trajectory for nonlinear systems. It is assumed that the reference trajectory is generated by means of a neutrally stable unforced system (exosystem) whose state is known. The design methodology is based on the center manifold theory and on the equivalent control concept and can be applied to both minimum and nonminimum-phase systems
Adaptive output regulation via post-processing internal models and hybrid identifiers
No full textThis paper deals with the problem of adaptive output regulation of single-input single-output nonlinear systems, with respect to uncertainties in the exosystem. We endow a recently developed post-processing internal model design with a hybrid adaptive structure, which allows to use different identification schemes to adaptively tune the internal model at runtime. Practical regulation results are presented, with the regulation error that is proved to be linearly related to the prediction capabilities of the identifier
- …
