This thesis explores the use of an Autonomous Underwater Vehicle (AUV) to track and pursue a tagged shark through the water. A controller was designed to take bearing and range to the shark tag and then control the AUV to pursue it. First, the ability of a particle filter to provide an accurate estimation of the location of the shark relative to the AUV is explored. Second, the ability of the AUV to follow the shark's path through the water is shown. This ability allows for localized environmental sampling of the shark's preferred path. Third, various path weightings are used to optimize the efficiency of pursuing the shark. This demonstrates that the proposed controller is efficient and effective. Fourth, the benefits of the addition of a second AUV are explored and quantified. The secondary AUV is shown to maintain formation without direct communication from the primary AUV. However, the communication of the AUVs increases the accuracy of all measurements and allows for future expansion in the complexity of the controller. Fifth, the effects of predicting the shark$'$s future movement is explored. Sixth, the effect of noise in the signal from the shark tag is tested and the level of noise at which the AUV can no longer pursue the shark is shown. This investigates the real world ability of the controller to accept noisy inputs and still generate the appropriate response. Finally, the positive results of the previous sections are combined and tested for various noise levels to show the improved controller response even under increased noise levels. To validate the proposed estimator and controller, seven tests were conducted. All tests were conducted on existing shark path data recorded by a stationary acoustic receiver and a boat mounted acoustic receiver. Tests were conducted on data sets from two different species of sharks, (Shovelnose and White) with two very different swimming behaviors. This shows the solution's flexibility in the species of shark tracked.

In this thesis the eigenstructure (eigenvalues and eigenvectors) assignment technique based algorithm has been developed for the design of controllers for power system applications. The application of the algorithm is demonstrated by designing power system stabilizers (PSSs) that are extensively used to address the small-signal rotor angle stability problems in power systems. In the eigenstructure assignment technique, the critical eigenvalues can be relocated as well as their associated eigenvectors can be modified. This method is superior and yield better dynamical performance compared to the widely used frequency domain design method, in which only the critical eigenvalues are relocated and no attempt is made to modify the eigenvectors. The reviewed published research has demonstrated successful application of the eigenstructure assignment technique in the design of controllers for small control systems. However, the application of this technique in the design of controllers for power systems has not been investigated rigorously. In contrast to a small system, a power system has a very large number state variables compared to the combined number of system inputs and outputs. Therefore, the eigenstructure assignment technique that has been successfully applied in the design of controllers for small systems could not be applied as is in the design of power system controllers. This thesis proposes a novel approach to the application of the eigenstructure assignment technique in the design of power system controllers. In this new approach, a multi-objective nonlinear optimization problem (MONLOP) is formulated by quantifying different design objectives as a function of free parametric vectors. Then the MONLOP is solved for the free parametric vectors using a nonlinear optimization technique. Finally, the solution of the controller parameters is obtained using the solved free parametric vectors. The superiority of the proposed method over the conventional frequency domain method is demonstrated by designing controllers for three different systems and validating the controllers through nonlinear transient simulations. One of the cases includes design of a PSS for the Manitoba Hydro system having about 29,000 states variables, which demonstrates the applicability of the proposed algorithm for a practical real-world system.

In this thesis the eigenstructure (eigenvalues and eigenvectors) assignment technique based algorithm has been developed for the design of controllers for power system applications. The application of the algorithm is demonstrated by designing power system stabilizers (PSSs) that are extensively used to address the small-signal rotor angle stability problems in power systems. In the eigenstructure assignment technique, the critical eigenvalues can be relocated as well as their associated eigenvectors can be modified. This method is superior and yield better dynamical performance compared to the widely used frequency domain design method, in which only the critical eigenvalues are relocated and no attempt is made to modify the eigenvectors. The reviewed published research has demonstrated successful application of the eigenstructure assignment technique in the design of controllers for small control systems. However, the application of this technique in the design of controllers for power systems has not been investigated rigorously. In contrast to a small system, a power system has a very large number state variables compared to the combined number of system inputs and outputs. Therefore, the eigenstructure assignment technique that has been successfully applied in the design of controllers for small systems could not be applied as is in the design of power system controllers. This thesis proposes a novel approach to the application of the eigenstructure assignment technique in the design of power system controllers. In this new approach, a multi-objective nonlinear optimization problem (MONLOP) is formulated by quantifying different design objectives as a function of free parametric vectors. Then the MONLOP is solved for the free parametric vectors using a nonlinear optimization technique. Finally, the solution of the controller parameters is obtained using the solved free parametric vectors. The superiority of the proposed method over the conventional frequency domain method is demonstrated by designing controllers for three different systems and validating the controllers through nonlinear transient simulations. One of the cases includes design of a PSS for the Manitoba Hydro system having about 29,000 states variables, which demonstrates the applicability of the proposed algorithm for a practical real-world system.

No description available.

Understanding and synthesizing locomotion of humans and animals will have far-reaching impacts in computer animation, robotic and biomechanics. However, due to the complexity of the neuromuscular control and physical interactions with the environment, computationally modeling these seemingly effortless locomotion imposes a grand challenge for scientists, engineers and artists. The focus of this thesis is to present a set of computational tools, which can simulate the physical environment and optimize the control strategy, to automatically synthesize locomotion for humans and animals. We first present computational tools to study swimming motions for a wide variety of aquatic animals. This method first builds a simulation of two-way interaction between fluid and an articulated rigid body system. It then searches for the most energy efficient way to swim for a given body shape in the simulated hydrodynamic environment. Next, we present an algorithm that can synthesize locomotion of soft body animals that do not have skeleton support. We combine a finite element simulation with a muscle model that is inspired by muscular hydrostat in nature. We then formulate a quadratic program with complementarity condition (QPCC) to optimize the muscle contraction and contact forces that can lead to meaningful locomotion. We develop an efficient QPCC solver that solves a challenging optimization problem at the presence of discontinuous contact events. We also present algorithms to model human locomotion with a passive mechanical device: riding a bicycle in this case. We apply a powerful reinforcement learning algorithm, which can search for both the parametrization and the parameters of a control policy, to enable a virtual human character to perform bicycle stunts in a physically simulated environment. Finally, we explore the possibility to use the computational tools that are developed for computer animation to control a real robot. We develop a simulation calibration technique which reduces the discrepancy between the simulated results and the performance of the robot in the real environments. For certain motion planning tasks, this method can transfer the controllers optimized for a virtual character in a simulation to a robot that operates in a real environment.

This thesis studies cloud capacity auto-scaling, or how to provision and release re-sources to a service running in the cloud based on its actual demand using an auto-matic controller. As the performance of server systems depends on the system design,the system implementation, and the workloads the system is subjected to, we focuson these aspects with respect to designing auto-scaling algorithms. Towards this goal,we design and implement two auto-scaling algorithms for cloud infrastructures. Thealgorithms predict the future load for an application running in the cloud. We discussthe different approaches to designing an auto-scaler combining reactive and proactivecontrol methods, and to be able to handle long running requests, e.g., tasks runningfor longer than the actuation interval, in a cloud. We compare the performance ofour algorithms with state-of-the-art auto-scalers and evaluate the controllers’ perfor-mance with a set of workloads. As any controller is designed with an assumptionon the operating conditions and system dynamics, the performance of an auto-scalervaries with different workloads.In order to better understand the workload dynamics and evolution, we analyze a6-years long workload trace of the sixth most popular Internet website. In addition,we analyze a workload from one of the largest Video-on-Demand streaming servicesin Sweden. We discuss the popularity of objects served by the two services, the spikesin the two workloads, and the invariants in the workloads. We also introduce, a mea-sure for the disorder in a workload, i.e., the amount of burstiness. The measure isbased on Sample Entropy, an empirical statistic used in biomedical signal processingto characterize biomedical signals. The introduced measure can be used to charac-terize the workloads based on their burstiness profiles. We compare our introducedmeasure with the literature on quantifying burstiness in a server workload, and showthe advantages of our introduced measure.To better understand the tradeoffs between using different auto-scalers with differ-ent workloads, we design a framework to compare auto-scalers and give probabilisticguarantees on the performance in worst-case scenarios. Using different evaluation cri-teria and more than 700 workload traces, we compare six state-of-the-art auto-scalersthat we believe represent the development of the field in the past 8 years. Knowingthat the auto-scalers’ performance depends on the workloads, we design a workloadanalysis and classification tool that assigns a workload to its most suitable elasticitycontroller out of a set of implemented controllers. The tool has two main components;an analyzer, and a classifier. The analyzer analyzes a workload and feeds the analysisresults to the classifier. The classifier assigns a workload to the most suitable elasticitycontroller based on the workload characteristics and a set of predefined business levelobjectives. The tool is evaluated with a set of collected real workloads, and a set ofgenerated synthetic workloads. Our evaluation results shows that the tool can help acloud provider to improve the QoS provided to the customers.

Commercial Buildings and complexes are no longer just national heat and power network energy loads, but they are becoming part of a smarter grid by including their own dedicated local heat and power generation. They do this by utilising both heat and power networks/micro-grids. A building integrated approach of Combined Heat and Power (CHP) generation with photovoltaic power generation (PV) abbreviated as CHPV is emerging as a complementary energy supply solution to conventional (i.e. national grid based) gas and electricity grid supplies in the design of sustainable commercial buildings and communities. The merits for the building user/owner of this approach are: to reduce life time energy running costs; reduce carbon emissions to contribute to UK’s 2020/2030 climate change targets; and provide a more flexible and controllable local energy system to act as a dynamic supply and/or load to the central grid infrastructure. The energy efficiency and carbon dioxide (CO2) reductions achievable by CHP systems are well documented. The merits claimed by these solutions are predicated on the ability of these systems being able to satisfy: perfect matching of heat and power supply and demand; ability at all times to maintain high quality power supply; and to be able to operate with these constraints in a highly dynamic and unpredictable heat and power demand situation. Any circumstance resulting in failure to guarantee power quality or matching of supply and demand will result in a degradation of the achievable energy efficiency and CO2 reduction. CHP based local energy systems cannot rely on large scale diversity of demand to create a relatively easy approach to supply and demand matching (i.e. as in the case of large centralised power grid infrastructures). The diversity of demand in a local energy system is both much greater than the centralised system and is also specific to the local system. It is therefore essential that these systems have robust and high performance control systems to ensure supply and demand matching and high power quality can be achieved at all times. Ideally this same control system should be able to make best use of local energy system energy storage to enable it to be used as a flexible, highly responsive energy supply and/or demand for the centralised infrastructure. In this thesis, a comprehensive literature survey has identified that there is no scientific and rigorous method to assess the controllability or the design of control systems for these local energy systems. Thus, the main challenge of the work described in this thesis is that of a controller design method and modelling approach for CHP based local energy systems. Specifically, the main research challenge for the controller design and modelling methodology was to provide an accurate and stable system performance to deliver a reliable tracking of power drawn/supplied to the centralised infrastructure whilst tracking the require thermal comfort in the local energy systems buildings. In the thesis, the CHPV system has been used as a case study. A CHPV based solution provides all the benefits of CHP combined with the near zero carbon building/local network integrated PV power generation. CHPV needs to be designed to provide energy for the local buildings’ heating, dynamic ventilating system and air-conditioning (HVAC) facilities as well as all electrical power demands. The thesis also presents in addition to the controller design and modelling methodology a novel CHPV system design topology for robust, reliable and high-performance control of building temperatures and energy supply from the local energy system. The advanced control system solution aims to achieve desired building temperatures using thermostatic control whilst simultaneously tracking a specified national grid power demand profile. The theory is innovative as it provides a stability criterion as well as guarantees to track a specified dynamic grid connection demand profile. This research also presents: design a dynamic MATLAB simulation model for a 5-building zone commercial building to show the efficacy of the novel control strategy in terms of: delivering accurate thermal comfort and power supply; reducing the amount of CO2 emissions by the entire energy system; reducing running costs verses national rid/conventional approaches. The model was developed by inspecting the functional needs of 3 local energy system case studies which are also described in the thesis. The CHPV system is combined with supplementary gas boiler for additional heating to guarantee simultaneous tracking of all the zones thermal comfort requirements whilst simultaneously tracking a specified national grid power demand using a Photovoltaics array to supply the system with renewable energy to reduce amount of CO2 emission. The local energy system in this research can operate in any of three modes (Exporting, Importing, Island). The emphasise of the thesis modelling method has been verified to be applicable to a wide range of case studies described in the thesis chapter 3. This modelling framework is the platform for creating a generic controlled design methodology that can be applied to all these case studies and beyond, including Local Energy System (LES) in hotter climates that require a cooling network using absorption chillers. In the thesis in chapter 4 this controller design methodology using the modelling framework is applied to just one case study of Copperas Hill. Local energy systems face two types of challenges: technical and nontechnical (such as energy economics and legislation). This thesis concentrates solely on the main technical challenges of a local energy system that has been identified as a gap in knowledge in the literature survey. The gap identified is the need for a controller design methodology to allow high performance and safe integration of the local energy system with the national grid infrastructure and locally installed renewables. This integration requires the system to be able to operate at high performance and safely in all different modes of operation and manage effectively the multi-vector energy supply system (e.g. simultaneous supply of heat and power from a single system).

Despite the vast amount of delivered theoretical results, regarding the topic of controller design, more than 90% of the controllers used in industry (petro-chemical, pulp and paper, steel, mining, etc) are of PID type (P, PI, PII, PD). This shows the importance of progressing in the elaboration of methods that consider restricted complexity controllers for practical applications, and that are computationally simple. Iterative Feedback Tuning (IFT) stands out as a new solution that takes into account both constraints. It belongs to the family of model-free controller tuning methods. It was developed at Cesame in the nineties and, since then, many real applications of IFT have been reported. This algorithm minimizes a cost function by means of a stochastic gradient descent scheme. In spite of the fact that the method has had an unexpected success in the tuning of real processes, a number of issues had not been fully covered yet. This thesis focuses on two aspects of this set of uncovered theoretical points: the convergence rate of the algorithm and a robust estimation of its gradient. Optimal prefilters, left as a degree of freedom for the user in the first formulation of IFT, are computed at each experiment. Their application allows a reduction in the covariance of the gradient estimate. Depending on what particular aspect the user is interested in improving, one optimal prefilter is selected. Monte-Carlo simulations have shown an enhancement with regards to a constant prefilter. A flexible arm set-up mounted in our robotics laboratory is used as a test bed to compare a model-based controller design algorithm with a model-free controller tuning method. The comparison is performed with some specifications defined beforehand. The same set-up plus a couple of air-jets serves as a tester for our theoretical results, when the rejection of a perturbation is the ultimate objective. Both cases have confirmed the predicted good behaviour offered by IFT.

A method of controller design for multiple input multiple output (MIMO) system is needed that will not give the high order controllers of modern control theory but will be more systematic than the “ad hoc” method. The objective of this method of design for multiple input multiple output systems is to find a controller of fixed order with performance specifications taken into consideration. An inner approximation of the stabilizing set is found through the algorithm discussed in Keel and Bhattacharyya’s "Fixed order multivariable controller synthesis: A new algorithm." The set satisfying the performance is then approximated through one of two algorithms; a hybrid of two optimization algorithms or the grid algorithm found in Lampton’s "Reinforcement Learning of a Morphing Airfoil-Policy and Discrete Learning Analysis." The method is then applied to five models of four aircraft; Commander 700, X-29, X-38, and F-5A using controllers of first and second orders.

This thesis presents the design and implementation of a single-stage with high power factor asymmetrical half-bridge converter. The main structure combines a boost converter with an asymmetrical half-bridge. An Asymmetrical half-bridge converter has many advantages such as soft-switching properties and fewer components. Therefore it is suitable for DC/DC cell. The boost converter is used in a PFC cell that operating in discontinuous condition mode have innate ability of power factor correction without additional controller. In this thesis, the complete analysis of operation principle and design of controller for the equivalent circuits of a single-stage AC/DC converter in every operating stage have been described in detail. Finally, we construct the single-stage circuit and experimental result show that it can reach the expected goal for power-factor correction.