Model-based control strategies for robot manipulators can present numerous performance advantages when an accurate model of the system dynamics is available. In practice, obtaining such a model is a challenging task which involves modeling such physical processes as friction, which may not be well understood and difficult to model. Furthermore, uncertainties in the physical parameters of a system may be introduced from significant discrepancies between the manufacturer data and the actual system. Traditionally, adaptive and robust control strategies have been developed to deal with parametric uncertainty in the dynamic model, but often require knowledge of the structure of the dynamics. Recent approaches to model-based manipulator control involve data-driven learning of the inverse dynamics relationship, eliminating the need for any a-priori knowledge of the system model. Locally Weighted Projection Regression (LWPR) has been proposed for learning the inverse dynamics function of a manipulator. Due to its use of simple local, linear models, LWPR is suitable for online and incremental learning. Although global regression techniques such as Gaussian Process Regression (GPR) have been shown to outperform LWPR in terms of accuracy, due to its heavy computational requirements, GPR has been applied mainly to offline learning of inverse dynamics. More recent efforts in making GPR computationally tractable for real-time control have resulted in several approximations which operate on a select subset, or sparse representation of the entire training data set. Despite the significant advancements that have been made in the area of learning control, there has not been much work in recent years to evaluate these newer regression techniques against traditional model-based control strategies such as adaptive control. Hence, the first portion of this thesis provides a comparison between a fixed model-based control strategy, an adaptive controller and the LWPR-based learning controller. Simulations are carried out in order to evaluate the position and orientation tracking performance of each controller under varied end effector loading, velocities and inaccuracies in the known dynamic parameters. Both the adaptive controller and LWPR controller are shown to have comparable performance in the presence of parametric uncertainty. However, it is shown that the learning controller is unable to generalize well outside of the regions in which it has been trained. Hence, achieving good performance requires significant amounts of training in the anticipated region of operation. In addition to poor generalization performance, most learning controllers commence learning entirely from `scratch,' making no use of any a-priori knowledge which may be available from the well-known rigid body dynamics (RBD) formulation. The second portion of this thesis develops two techniques for online, incremental learning algorithms which incorporate prior knowledge to improve generalization performance. First, prior knowledge is incorporated into the LWPR framework by initializing the local linear models with a first order approximation of the prior information. Second, prior knowledge is incorporated into the mean function of Sparse Online Gaussian Processes (SOGP) and Sparse Pseudo-input Gaussian Processes (SPGP), and a modified version of the algorithm is proposed to allow for online, incremental updates. It is shown that the proposed approaches allow the system to operate well even without any initial training data, and further performance improvement can be achieved with additional online training. Furthermore, it is also shown that even partial knowledge of the system dynamics, for example, only the gravity loading vector, can be used effectively to initialize the learning.

No description available.

<p>Good models of industrial robots are necessary in a variety of applications, such as mechanical design, performance simulation, control, diagnosis, supervision and offline programming. This motivates the need for good modelling tools. In the first part of this thesis the forward kinematic modelling of serial industrial robots is studied. The first steps towards a toolbox are implemented in the Maple programming language.</p><p>A series of possible applications for the toolbox can be mentioned. One example is to estimate the pose of the robot tool using an extended Kalman filter by means of extra sensors mounted on the robot. The kinematic equations and the relations necessary for the extended Kalman filter can be derived in the modelling tool. Iterative learning control, ILC, using an estimate of the tool position can then improve the robot performance.</p><p>The second part of the thesis is devoted to ILC, which is a control method that is applicable when the robot performs a repetitive movement starting from the same initial conditions every repetition. The algorithm compensates for repetitive errors by adding a correction signal to the reference. Studies where ILC is applied to a real industrial platform is less common in the literature, which motivates the work in this thesis.</p><p>A first-order ILC filter with iteration-independent operators derived using a heuristic design approach is used, which results in a non-causal algorithm. A simulation study is made, where a flexible two-mass model is used as a simplified linear model of a single robot joint and the ILC algorithm applied is based on motor-angle measurements only. It is shown that when a model error is introduced in the relation between the arm and motor reference angle, it is not necessary that the error on the arm side is reduced as much as the error on the motor side, or in fact reduced at all.</p><p>In the experiments the ILC algorithm is applied to a large-size commercial industrial robot, performing a circular motion that is relevant for a laser-cutting application. The same ILC design variables are used for all six motors and the learning is stopped after five iterations, which is motivated in practice by experimental results. Performance on the motor side and the corresponding performance on the arm side, using a laser-measurement system, is studied. Even though the result on the motor side is good, it is no guarantee that the errors on the arm side are decreasing. One has to be very careful when dealing with resonant systems when the controlled variable is not directly measured and included in the algorithm. This indicates that the results on the arm side may be improved when an estimate of, for example, the tool position is used in the ILC algorithm.</p> / <p>Bra modeller av industrirobotar behövs i en mängd olika tillämpningar, som till exempel mekanisk design, simulering av prestanda, reglering, diagnos, övervakning och offline-programmering. I första delen av avhandlingen studeras modellering av framåtkinematiken för en seriell robot och implementeringen av ett modelleringsverktyg, en toolbox, för kinematikmodellering i Maple beskrivs ingående.</p><p>Ett antal möjliga tillämpningar för toolboxen kan nämnas. Ett exempel är att med hjälp av extra sensorer monterade på roboten och ett så kallat extended Kalmanfilter förbättra skattningen av positionen och orienteringen för robotverktyget. De kinematiska ekvationerna och sambanden som behövs för extended Kalmanfiltret kan beräknas med hjälp av modelleringsverktyget. Reglering genom iterativ inlärning - iterative learning control, ILC - där en skattning av verktygspositionen används, kan sedan förbättra robotens prestanda.</p><p>Andra delen av avhandlingen är tillägnad ILC. Det är en reglermetod som är användbar när roboten utför en repetitiv rörelse som startar från samma initialvillkor varje gång. Algoritmen kompenserar för de repetitiva felen genom att addera en korrektionsterm till referenssignalen. Studier där ILC är tillämpad på en verklig industriell plattform är mindre vanligt i litteraturen, vilket motiverar arbetet i avhandlingen.</p><p>Ett första ordningens ILC-filter med iterationsoberoende operatorer används. ILC-algoritmen är framtagen enligt ett heuristiskt tillvägagångssätt, vilket resulterar i en ickekausal algoritm. I en simuleringsstudie med en flexibel tvåmassemodell som en förenklad linjär modell av en enskild robotled, används en ILC-algoritm baserad endast på motorvinkelmätningar. Det visar sig att när ett modellfel introduceras i sambandet mellan arm- och motorvinkelreferensen, är det inte säkert att felet på armsidan minskar så mycket som felet på motorsidan, eller minskar överhuvudtaget.</p><p>I experiment tillämpas ILC-algoritmen på en stor kommersiell industrirobot som utför en cirkelrörelse som är relevant för en laserskärningstillämpning. Samma designvariabler används för alla sex motorerna och inlärningen stoppas efter fem iterationer, vilket är motiverat i praktiken genom experimentella resultat. Prestanda på motorsidan studeras, och motsvarande prestanda på armsidan mäts med ett lasermätsystem. Trots goda resultat på motorsidan finns det inga garantier för minskande fel på armsidan. Stor försiktighet krävs när experimenten innefattar ett resonant system där den reglerade variabeln inte är mätt explicit och inkluderad i algoritmen. Detta visar på möjligheten att förbättra resultaten på armsidan då en skattning av till exempel verktygspositionen används i ILC-algoritmen.</p> / Report code: LiU-TEK-LIC-2008:1.

No description available.

This research addresses the design and development of a decision support tool to provide healthcare policy makers with insights and feedback when evaluating proposed patient flow and infection mitigation and control strategies in the emergency department (ED). An agent-based modeling (ABM) approach was used to simulate EDs, designed to be tuneable to specific parameters related to specification of topography, agent characteristics and behaviours, and the application in question. In this way, it allows for the user to simulate various ‘what-if’ scenarios related to infection spread and patient flow, where such policy questions may otherwise be left “best intent open loop” in practice. Infection spread modeling and patient flow modeling have been addressed by mathematical and queueing models in the past; however, the application of an ABM approach at the level of an institution is novel. A conjecture of this thesis is that such a tool should be augmented with Machine Learning (ML) technology to assist in performing optimization or search in which patient flow and infection spread are signals or variables of interest. Therefore this work seeks to design and demonstrate a decision support tool with ML capability for optimizing ED processes. The primary contribution of this thesis is the development of a novel, flexible, and tuneable framework for spatial, human-scale ABM in the context of a decision support tool for healthcare policy relating to infection spread and patient flow within EDs . The secondary contribution is the demonstration of the utility of ML for automatic policy generation with respect to the ABM tool. The application of ML to automatically generate healthcare policy in concert with an ABM is believed to be novel and of emerging practical importance. The tertiary contribution is the development and testing of a novel heuristic specific to the ML paradigm used: Genetic Programming (GP). This heuristic aids learning tasks performed in conjunction with ABMs for healthcare policy. The primary contribution is clearly demonstrated within this thesis. The others are of a more difficult nature; the groundwork has been laid for further work in these areas that are likely to remain open for the foreseeable future.

This research addresses the design and development of a decision support tool to provide healthcare policy makers with insights and feedback when evaluating proposed patient flow and infection mitigation and control strategies in the emergency department (ED). An agent-based modeling (ABM) approach was used to simulate EDs, designed to be tuneable to specific parameters related to specification of topography, agent characteristics and behaviours, and the application in question. In this way, it allows for the user to simulate various ‘what-if’ scenarios related to infection spread and patient flow, where such policy questions may otherwise be left “best intent open loop” in practice. Infection spread modeling and patient flow modeling have been addressed by mathematical and queueing models in the past; however, the application of an ABM approach at the level of an institution is novel. A conjecture of this thesis is that such a tool should be augmented with Machine Learning (ML) technology to assist in performing optimization or search in which patient flow and infection spread are signals or variables of interest. Therefore this work seeks to design and demonstrate a decision support tool with ML capability for optimizing ED processes. The primary contribution of this thesis is the development of a novel, flexible, and tuneable framework for spatial, human-scale ABM in the context of a decision support tool for healthcare policy relating to infection spread and patient flow within EDs . The secondary contribution is the demonstration of the utility of ML for automatic policy generation with respect to the ABM tool. The application of ML to automatically generate healthcare policy in concert with an ABM is believed to be novel and of emerging practical importance. The tertiary contribution is the development and testing of a novel heuristic specific to the ML paradigm used: Genetic Programming (GP). This heuristic aids learning tasks performed in conjunction with ABMs for healthcare policy. The primary contribution is clearly demonstrated within this thesis. The others are of a more difficult nature; the groundwork has been laid for further work in these areas that are likely to remain open for the foreseeable future.

This thesis focuses on the theory, design, simulation and implementation of several digital controllers for periodic signals on a laser scanning galvanometer. A model for the galvanometer was obtained and veri ed using closed loop identi cation techniques. Using this model, controllers were designed and simulated using MATLAB and then implemented on a custom FPGA control processor with a focus on tracking performance. The types of controllers used were: an Iterative Learning Controller, an RST pole placement controller, an Adaptive Feed-forward cancellation controller, a combined Iterative Learning and Adaptive Feed-forward cancellation controller and a simple PID controller. The simulated results were better than the experimental results because of system noise and modelling uncertainties but the relative performance between each of the controllers was similar for both the simulation and experimental setup. The experimental results achieved were very good with one controller reaching errors under 50 rad. / Graduate / 0537 / t_Torres_bonet@hotmail.com

Model-based control strategies for robot manipulators can present numerous performance advantages when an accurate model of the system dynamics is available. In practice, obtaining such a model is a challenging task which involves modeling such physical processes as friction, which may not be well understood and difficult to model. Furthermore, uncertainties in the physical parameters of a system may be introduced from significant discrepancies between the manufacturer data and the actual system. Traditionally, adaptive and robust control strategies have been developed to deal with parametric uncertainty in the dynamic model, but often require knowledge of the structure of the dynamics. Recent approaches to model-based manipulator control involve data-driven learning of the inverse dynamics relationship, eliminating the need for any a-priori knowledge of the system model. Locally Weighted Projection Regression (LWPR) has been proposed for learning the inverse dynamics function of a manipulator. Due to its use of simple local, linear models, LWPR is suitable for online and incremental learning. Although global regression techniques such as Gaussian Process Regression (GPR) have been shown to outperform LWPR in terms of accuracy, due to its heavy computational requirements, GPR has been applied mainly to offline learning of inverse dynamics. More recent efforts in making GPR computationally tractable for real-time control have resulted in several approximations which operate on a select subset, or sparse representation of the entire training data set. Despite the significant advancements that have been made in the area of learning control, there has not been much work in recent years to evaluate these newer regression techniques against traditional model-based control strategies such as adaptive control. Hence, the first portion of this thesis provides a comparison between a fixed model-based control strategy, an adaptive controller and the LWPR-based learning controller. Simulations are carried out in order to evaluate the position and orientation tracking performance of each controller under varied end effector loading, velocities and inaccuracies in the known dynamic parameters. Both the adaptive controller and LWPR controller are shown to have comparable performance in the presence of parametric uncertainty. However, it is shown that the learning controller is unable to generalize well outside of the regions in which it has been trained. Hence, achieving good performance requires significant amounts of training in the anticipated region of operation. In addition to poor generalization performance, most learning controllers commence learning entirely from `scratch,' making no use of any a-priori knowledge which may be available from the well-known rigid body dynamics (RBD) formulation. The second portion of this thesis develops two techniques for online, incremental learning algorithms which incorporate prior knowledge to improve generalization performance. First, prior knowledge is incorporated into the LWPR framework by initializing the local linear models with a first order approximation of the prior information. Second, prior knowledge is incorporated into the mean function of Sparse Online Gaussian Processes (SOGP) and Sparse Pseudo-input Gaussian Processes (SPGP), and a modified version of the algorithm is proposed to allow for online, incremental updates. It is shown that the proposed approaches allow the system to operate well even without any initial training data, and further performance improvement can be achieved with additional online training. Furthermore, it is also shown that even partial knowledge of the system dynamics, for example, only the gravity loading vector, can be used effectively to initialize the learning.

Good models of industrial robots are necessary in a variety of applications, such as mechanical design, performance simulation, control, diagnosis, supervision and offline programming. This motivates the need for good modelling tools. In the first part of this thesis the forward kinematic modelling of serial industrial robots is studied. The first steps towards a toolbox are implemented in the Maple programming language. A series of possible applications for the toolbox can be mentioned. One example is to estimate the pose of the robot tool using an extended Kalman filter by means of extra sensors mounted on the robot. The kinematic equations and the relations necessary for the extended Kalman filter can be derived in the modelling tool. Iterative learning control, ILC, using an estimate of the tool position can then improve the robot performance. The second part of the thesis is devoted to ILC, which is a control method that is applicable when the robot performs a repetitive movement starting from the same initial conditions every repetition. The algorithm compensates for repetitive errors by adding a correction signal to the reference. Studies where ILC is applied to a real industrial platform is less common in the literature, which motivates the work in this thesis. A first-order ILC filter with iteration-independent operators derived using a heuristic design approach is used, which results in a non-causal algorithm. A simulation study is made, where a flexible two-mass model is used as a simplified linear model of a single robot joint and the ILC algorithm applied is based on motor-angle measurements only. It is shown that when a model error is introduced in the relation between the arm and motor reference angle, it is not necessary that the error on the arm side is reduced as much as the error on the motor side, or in fact reduced at all. In the experiments the ILC algorithm is applied to a large-size commercial industrial robot, performing a circular motion that is relevant for a laser-cutting application. The same ILC design variables are used for all six motors and the learning is stopped after five iterations, which is motivated in practice by experimental results. Performance on the motor side and the corresponding performance on the arm side, using a laser-measurement system, is studied. Even though the result on the motor side is good, it is no guarantee that the errors on the arm side are decreasing. One has to be very careful when dealing with resonant systems when the controlled variable is not directly measured and included in the algorithm. This indicates that the results on the arm side may be improved when an estimate of, for example, the tool position is used in the ILC algorithm. / Bra modeller av industrirobotar behövs i en mängd olika tillämpningar, som till exempel mekanisk design, simulering av prestanda, reglering, diagnos, övervakning och offline-programmering. I första delen av avhandlingen studeras modellering av framåtkinematiken för en seriell robot och implementeringen av ett modelleringsverktyg, en toolbox, för kinematikmodellering i Maple beskrivs ingående. Ett antal möjliga tillämpningar för toolboxen kan nämnas. Ett exempel är att med hjälp av extra sensorer monterade på roboten och ett så kallat extended Kalmanfilter förbättra skattningen av positionen och orienteringen för robotverktyget. De kinematiska ekvationerna och sambanden som behövs för extended Kalmanfiltret kan beräknas med hjälp av modelleringsverktyget. Reglering genom iterativ inlärning - iterative learning control, ILC - där en skattning av verktygspositionen används, kan sedan förbättra robotens prestanda. Andra delen av avhandlingen är tillägnad ILC. Det är en reglermetod som är användbar när roboten utför en repetitiv rörelse som startar från samma initialvillkor varje gång. Algoritmen kompenserar för de repetitiva felen genom att addera en korrektionsterm till referenssignalen. Studier där ILC är tillämpad på en verklig industriell plattform är mindre vanligt i litteraturen, vilket motiverar arbetet i avhandlingen. Ett första ordningens ILC-filter med iterationsoberoende operatorer används. ILC-algoritmen är framtagen enligt ett heuristiskt tillvägagångssätt, vilket resulterar i en ickekausal algoritm. I en simuleringsstudie med en flexibel tvåmassemodell som en förenklad linjär modell av en enskild robotled, används en ILC-algoritm baserad endast på motorvinkelmätningar. Det visar sig att när ett modellfel introduceras i sambandet mellan arm- och motorvinkelreferensen, är det inte säkert att felet på armsidan minskar så mycket som felet på motorsidan, eller minskar överhuvudtaget. I experiment tillämpas ILC-algoritmen på en stor kommersiell industrirobot som utför en cirkelrörelse som är relevant för en laserskärningstillämpning. Samma designvariabler används för alla sex motorerna och inlärningen stoppas efter fem iterationer, vilket är motiverat i praktiken genom experimentella resultat. Prestanda på motorsidan studeras, och motsvarande prestanda på armsidan mäts med ett lasermätsystem. Trots goda resultat på motorsidan finns det inga garantier för minskande fel på armsidan. Stor försiktighet krävs när experimenten innefattar ett resonant system där den reglerade variabeln inte är mätt explicit och inkluderad i algoritmen. Detta visar på möjligheten att förbättra resultaten på armsidan då en skattning av till exempel verktygspositionen används i ILC-algoritmen. / <p>Report code: LiU-TEK-LIC-2008:1.</p>

The work in this dissertation concerns the construction of a robust iterative learning control (ILC) algorithm for a class of systems characterized by measurement delays, parametric uncertainty, and linear parameter varying (LPV) dynamics. One example of such a system is the twin roll strip casting process, which provides a practical motivation for this research. I propose three ILC algorithms in this dissertation that advance the state of the art. The first algorithm compensates for measurement delays that are longer than a single iteration of a periodic process. I divide the delay into an iterative and residual component and show how each component effects the asymptotic stability properties of the ILC algorithm. The second algorithm is a coupled delay estimation and ILC algorithm that compensates for time-varying measurement delays. I use an adaptive delay estimation algorithm to force the delay estimate to converge to the true delay and provide stability conditions for the coupled delay estimation and ILC algorithm. The final algorithm is a norm optimal ILC algorithm that compensates for LPV dynamics as well as parametric uncertainty and time delay estimation error. I provide a tuning method for the cost function weight matrices based on a sufficient condition for robust convergence and an upper bound on the norm of the error signal. The functionality of all three algorithms is demonstrated through simulated case studies based on an identified system model of the the twin roll strip casting process. The simulation testing is also augmented with experimental testing of select algorithms through collaboration with an industrial sponsor.