No description available.

Street lighting is an important aspect of infrastructure in terms of both safety and comfort, but it also consumes a lot of energy. Unused light is a waste of energy, and without any form of control of the street lighting, this problem will continue to increase along with the expansion of road networks. The aim of this thesis is to propose an intelligent control system for street lighting that can adapt to the velocity of individual road users, to investigate if this could provide ways to improve the efficiency of street lighting. Previous control approaches include systems based on ambient light intensity or presence of road users, but no studies were found in which illumination adapts to the velocity of road users. The project involves three main steps, including a literature review, a system implementation and evaluation. In the proposed system, street lights cooperate to detect road users and calculate their velocities in order to adapt the illumination and make it follow their movement. It can be concluded from the evaluation results that the velocity readings help further optimize the illumination control in comparison to systems that do not consider velocity. The velocity readings make it possible to only illuminate the roadway in the direction of travel, while also adapting the distance of illumination to the recorded speed. The proposed control scheme is considered a viable solution for reducing the amount of unused light, consequently reducing the energy consumption of street lighting.

Cooperative control of networked systems (or multi-agent systems) has attracted much attention during the past few years. But most of the existing results focus on first order and second order leaderless consensus problems with linear dynamics. The second part of this dissertation solves a higher-order synchronization problem for cooperative nonlinear systems with an active leader. The communication network considered is a weighted directed graph with fixed topology. Each agent is modeled by a higher-order nonlinear system with the nonlinear dynamics unknown. External unknown disturbances perturb each agent. The leader agent is modeled as a higher-order non-autonomous nonlinear system. It acts as a command generator and can only give commands to a small portion of the networked group. A robust adaptive neural network controller is designed for each agent. Neural network learning algorithms are given such that all nodes ultimately synchronize to the leader node with a small residual error. Moreover, these controllers are totally distributed in the sense that each controller only requires its own information and its neighbors' information. / Receding horizon control (RHC), also called model predictive control (MPC), is a suboptimal control scheme over an infinite horizon that is determined by solving a finite horizon open-loop optimal control problem repeatedly. It has widespread applications in industry. Reinforcement learning (RL) is a computational intelligence method in which an optimal control policy is learned over time by evaluating the performance of suboptimal control policies. In this dissertation it is shown that reinforcement learning techniques can significantly improve the behavior of RHC. Specifically, RL methods are used to add a learning feature to RHC. It is shown that keeping track of the value learned at the previous iteration and using it as the new terminal cost for RHC can overcome traditional strong requirements for RHC stability, such as that the terminal cost be a control Lyapunov function, or that the horizon length be greater than some bound. We propose improved RHC algorithms, called updated terminal cost receding horizon control (UTC-RHC), first in the framework of discrete-time linear systems and then in the framework of continuous-time linear systems. For both cases, we show the uniform exponential stability of the closed-loop system can be guaranteed under very mild conditions. Moreover, unlike RHC, the UTC-RHC control gain approaches the optimal policy associated with the infinite horizon optimal control problem. To show these properties, non-standard Lyapunov functions are introduced for both discrete-time case and continuous-time case. / Two topics of modern control are investigated in this dissertation, namely receding horizon control (RHC) and cooperative control of networked systems. We apply learning techniques to these two topics. Specifically, we incorporate the reinforcement learning concept into the standard receding horizon control, yielding a new RHC algorithm, and relax the stability constraints required for standard RHC. For the second topic, we apply neural adaptive control in synchronization of the networked nonlinear systems and propose distributed robust adaptive controllers such that all nodes synchronize to a leader node. / Zhang, Hongwei. / Adviser: Jie Huang. / Source: Dissertation Abstracts International, Volume: 72-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 99-105). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Thesis--Virginia Polytechnic Institute and State University, 1978. / Also available via the Internet.

Thesis (Ph. D.)--Ohio State University, 2006. / Title from first page of PDF file. Includes bibliographical references (p. 172-178).

To obtain a suitable mathematical model of the input-output behavior of highly nonlinear, multi-scale, nonparametric phenomena, we introduce an adaptive radial basis function approximation approach. We use this approach to estimate the discrepancy between traditional model areas and the multiscale physics of systems involving distributed sensing and technology. Radial Basis Function Networks offers the possible approach to nonparametric multi-scale modeling for dynamical systems like the adaptive wing with the Synthetic Jet Actuator (SJA). We use the Regularized Orthogonal Least Square method (Mark, 1996) and the RAN-EKF (Resource Allocating Network-Extended Kalman Filter) as a reference approach. The first part of the algorithm determines the location of centers one by one until the error goal is met and regularization is achieved. The second process includes an algorithm for the adaptation of all the parameters in the Radial Basis Function Network, centers, variances (shapes) and weights. To demonstrate the effectiveness of these algorithms, SJA wind tunnel data are modeled using this approach. Good performance is obtained compared with conventional neural networks like the multi layer neural network and least square algorithm. Following this work, we establish Model Reference Adaptive Control (MRAC) formulations using an off-line Radial Basis Function Networks (RBFN). We introduce the adaptive control law using a RBFN. A theory that combines RBFN and adaptive control is demonstrated through the simple numerical simulation of the SJA wing. It is expected that these studies will provide a basis for achieving an intelligent control structure for future active wing aircraft.

An improved robot manipulator decentralized non-linear adaptive controller that performs well in the presence of disturbances with unknown parameters and non-linearities is presented in this work. The proposed decentralized adaptive structure is a modification of the controller developed by Seraji [13-17] and is characterized by an auxiliary signal that compensates for the unmodeled dynamics and improves the tracking performance, by a feedforward component based on the inverse system to ensure high performance over a wide range and by a PD feedback component of constant gain to improve the speed of response of the system. As a result, a very accurate and fast path tracking is achieved despite the non-linearities. The scheme requires only the measurement of angular speed and displacement of each joint, and it does not require any knowledge about the mathematical model of the manipulator. Due to its decentralized structure, it can be implemented on parallel processors to speed up the operation. The main advantages of the proposed control scheme over similar controllers are that the control activity is smoother, it is less sensitive to sampling size and to the time period elapsed when the whole trajectory is traversed, as verified by simulations of several test conditions of-two of the joints of the PUMA 560 robot arm. / Graduation date: 1991

El disseny d'un controlador es un procés que requereix adquirir y processar informació amb la finalitat de dissenyar un sistema de control satisfactori. A més a més és àmpliament reconegut el fet de que si s'afegeix nova informació en el procés de disseny d'un controlador, és possible millorar el funcionament del controlador obtingut. Aquesta és la filosofia existent darrere del control adaptatiu. No obstant el concepte d'informació referit al problema del disseny de reguladors, si bé està molt estes, no disposa d'una formalització clara ni unificadora degut a la mancança d'un marc conceptual. Aquesta tesi aborda el problema del disseny de controladors des d'un punt de vista de la informació requerida per aconseguir aquesta finalitat. Les aportacions de la tesi es divideixen en dues parts. En la primera part l'objectiu és caracteritzar el concepte d'informació dins del problema del disseny de controladors, així com analitzar totes les fonts d'informació disponibles. Aquest objectiu s'aconsegueix mitjançant el desenvolupament d'un marc conceptual en el qual es pot establir relacions fonamentals necessàries per augmentar la informació dels elements del problema de control. En un segon pas, aquest marc ja establert s'utilitza par analitzar i comparar les tècniques del control adaptatiu clàssic amb el control iteratiu. El marc permet comparar ambdues tècniques de disseny de controladors en funció de la gestió de la informació que cadascuna d'elles realitza, proporcionant així un punt de referència per comparar diverses maneres de gestionar la informació per al disseny de controladors. En la segona part de la tesi s'aborda el problema de la validació de la informació existent en un model. L'objectiu és ser capaç de validar un model de manera que el resultat de la validació no sigui simplement un resultat binari de "validat/invalidat", si no que es donen guies de decisió sobre com gestionar la informació dels elements del problema de control amb la finalitat d'augmentar la informació dels models. Per aconseguir aquest fi es desenvolupa l'algorisme de validació FDMV (Frequency Domain Model Validation) que permet que el resultat de la validació d'un model sigui dependent de la freqüència. D'aquest fet es conclou que un mateix model pot ser validat per a un cert rang de freqüències mentre que el mateix model pot ser invalidat per a un altre rang de freqüències diferents. Aquesta validació dependent de la freqüència permet millorar la gestió de la informació en lo referent a 1) disseny experimental per a una nova identificació, 2) selecció de l'ordre del model adequat i, 3) selecció de la amplada de banda del controlador acceptable per l'actual model. L'algorisme FDMV es mostra com una eina especialment apropiada per a ser emprada en tècniques de control iteratiu. / El diseño de un controlador es un proceso que requiere adquirir y procesar información con el fin de diseñar un sistema de control satisfactorio. Además es ampliamente reconocido el hecho de que si se añade nueva información en el proceso de diseño de un controlador, es posible mejorar el desempeño del controlador obtenido. Esta es la filosofía existente detrás del control adaptativo. Sin embargo el concepto de información referido al problema de diseño de reguladores, si bien está muy extendido, no dispone de una formalización clara ni unificada al carecer de un marco conceptual. En esta tesis se aborda el problema del diseño de controladores desde un punto de vista de la información requerida para tal fin. Las aportaciones de la tesis se dividen en dos partes. En la primera parte el objetivo es caracterizar el concepto de información cuando hablamos del problema del diseño de controladores, así como analizar todas las fuentes de información disponibles. Esto se consigue mediante el desarrollo de un marco conceptual en el cual se pueden establecer relaciones fundamentales necesarias para aumentar la información de los elementos del problema de control. En un segundo paso, dicho marco ya establecido se utiliza para analizar y comparar las técnicas del control adaptativo clásico con el control iterativo. El marco permite comparar ambas técnicas de diseño de controladores en función de la gestión de la información que cada una de ellas realiza, proporcionando así un punto de referencia para comparar diversas maneras de gestionar la información para el diseño de controladores. En la segunda parte de la tesis se aborda el problema de la validación de la información existente en un modelo. El objetivo es ser capaz de validar un modelo de manera tal que el resultado de la validación no sea simplemente un resultado binario de "validado/invalidado", si no que aporte guías de decisión sobre como gestionar la información de los elementos del problema de control con el fin de aumentar la información del modelo. Para tal fin se desarrolla el algoritmo de validación FDMV (Frequency Domain Model Validation) que permite que el resultado de la validación de un modelo sea dependiente de la frecuencia. De ello se sigue que un mismo modelo puede ser validado para cierto rango de frecuencias mientras que el mismo modelo puede ser invalidado para otro rango de frecuencias diferentes. Esta validación dependiente de la frecuencia permite mejorar la gestión de la información en lo referente al i) diseño experimental para una nueva identificación, ii) selección del orden del modelo adecuado y iii) selección del ancho de banda del controlador aceptable por el modelo a mano. El algoritmo FDMV se muestra como una herramienta especialmente apropiada para ser empleada en técnicas de control iterativo.

A novel adaptive controller, suitable for linear and non-linear systems was developed. The controller is a discrete algorithm suitable for computer implementation and is based on gradient descent adaptation rules. Traditional recursive least squares based algorithms suffer from performance deterioration due to the continuous reduction of a covariance matrix used for adaptation. When this covariance matrix becomes too small, recursive least squares algorithms respond slow to changes in model parameters. Gradient descent adaptation was used to avoid the performance deterioration with time associated with regression based adaptation such as Recursive Least Squares methods. Stability was proven with Lyapunov stability theory, using an error filter designed to fulfill stability requirements. Similarities between the proposed controller with PI control have been found. A framework for on-line tuning was developed using the concept of estimation tracks. Estimation tracks allow the estimation gains to be selected from a finite set of possible values, while meeting Lyapunov stability requirements. The trade-off between sufficient excitation for learning and controller performance, typical for dual adaptive control techniques, are met by properly tuning the adaptation and filter gains to drive the rate of adaptation in response to a fixed excitation signal. Two methods for selecting the estimation track were developed. The first method uses simulations to predict the value of the bicriteria cost function that is a combination of prediction and feedback errors, to generate a performance score for each estimation track. The second method uses a linear matrix inequality formulation to find an upper bound on feedback error within the range of uncertainty of the plant parameters and acceptable reference signals. The linear matrix inequality approach was derived from a robust control approach. Numerical simulations were performed to systematically evaluate the performance and computational burden of configuration parameters, such as the number of estimation tracks used for tuning. Comparisons were performed for both tuning methods with an arbitrarily tuned adaptive controller, with arbitrarily selected tuning parameters as well as a common adaptive control algorithm.

A novel adaptive controller, suitable for linear and non-linear systems was developed. The controller is a discrete algorithm suitable for computer implementation and is based on gradient descent adaptation rules. Traditional recursive least squares based algorithms suffer from performance deterioration due to the continuous reduction of a covariance matrix used for adaptation. When this covariance matrix becomes too small, recursive least squares algorithms respond slow to changes in model parameters. Gradient descent adaptation was used to avoid the performance deterioration with time associated with regression based adaptation such as Recursive Least Squares methods. Stability was proven with Lyapunov stability theory, using an error filter designed to fulfill stability requirements. Similarities between the proposed controller with PI control have been found. A framework for on-line tuning was developed using the concept of estimation tracks. Estimation tracks allow the estimation gains to be selected from a finite set of possible values, while meeting Lyapunov stability requirements. The trade-off between sufficient excitation for learning and controller performance, typical for dual adaptive control techniques, are met by properly tuning the adaptation and filter gains to drive the rate of adaptation in response to a fixed excitation signal. Two methods for selecting the estimation track were developed. The first method uses simulations to predict the value of the bicriteria cost function that is a combination of prediction and feedback errors, to generate a performance score for each estimation track. The second method uses a linear matrix inequality formulation to find an upper bound on feedback error within the range of uncertainty of the plant parameters and acceptable reference signals. The linear matrix inequality approach was derived from a robust control approach. Numerical simulations were performed to systematically evaluate the performance and computational burden of configuration parameters, such as the number of estimation tracks used for tuning. Comparisons were performed for both tuning methods with an arbitrarily tuned adaptive controller, with arbitrarily selected tuning parameters as well as a common adaptive control algorithm.