The safety requirements for vehicles are today high and they will become more stringent in the future. The car companies test their products every day to ensure that safety requirements are met. These tests are often done by professional drivers. If the car is tested in an everyday traffic situation, a normal experienced driver is desired. A drawback is that a human will eventually learn the manoeuvre he/she is told to do. An artificial driver is therefore to prefer to make the test repeatable. This thesis’ purpose is to develop and implement an artificial driver as a controller in order to follow a predefined trajectory. The driver model’s performance driving a double lane change manoeuvre should be as close to a real driver’s as possible. Data was gathered by inviting people to drive in a simulator. The results from the simulator tests were used to implement three different drivers with different experiences. The gathered data was used to categorize the test drivers into different driver types for each specific velocity by using the vehicle position from thetest results. This thesis studies the driver from a controller’s perspective and it resulted in two implemented controllers for reference tracking. The first approach was a Model Predictive Controller with reference tracking and the other approach was to use a FIR-filter in order to describe the drivers’ characteristics. A vehicle model was implemented in order to do the double lane change manoeuvre in a simulation environment together with the implemented driver model. The results show that the two approaches can be used for reference tracking. The MPC showed good results with the recreation of the test runs that were made by the categorized drivers. The FIR-filter had problems to mimic the drivers’ test runs and their characteristics. The advantage with MPC is its robustness, while the advantages with the FIR-filter are its, in comparison, simplicity in the implementation and the algorithm’s low computational cost. In order to make the FIR-filter more robust, some improvements have to be made. One improvement is to use gain scheduling in order to adjust the filter coefficients depending on thevelocity. / De säkerhetskraven som idag ställs på fordon är höga och det kommer bli mer strikt i framtiden. Bilföretag testar sina bilar varje dag för att se om komponenterna och bilen klarar säkerhetskraven som ställs. Till dessa tester används professionella testförare. I en vardaglig trafiksituation är det önskvärt att en normalt erfaren bilförare utför testen. En mänsklig förare kommer använda sin inlärningsförmåga vid repeterande manöver, vilket inte är önskvärt. En artificiell förare är därför att föredra. Den artificiella föraren ska köra så likt en verklig förare som möjligt vid en "double lane change"- (DLC) manöver. Detta examensarbete har som avsikt att implementera en förare som en regulator för att kunna följa en förutbestämd trajektoria på samma sätt som en verklig förare. I detta examensarbete har "DLC"-manövern studerats. I examensarbetet har insamlad data från testförare använts för att kunna implementera tre olika förartyper med olika erfarenheter. Den insamlade datan användes till att kategorisera testförarna för varje särskild hastighet. Två tillvägagångssätt har gjorts med föraren, en där föraren är en modellbaserad prediktionsregulator med referensignalsföljning (MPC) och en där föraren implementeras som ett ändligt impulssvarsfilter (FIR-filter). En fordonsmodell har implementerats för att en "DLC"-manöver ska kunna testas i en simuleringsmiljö. Resultaten blev att de två metoderna klarade av referensföljningen. MPC:n var bra på att återskapa testförararnas körningar. FIR-filtret hade problem med att härma förarnas körningar och deras karaktäristik. Fördelen med MPC är dessrobusthet och fördelen med FIR-filtret är dess, i jämförelse, simplicitet vid implementering samt den låga beräkningskostnaden för algoritmen. För att göra FIR-filtret mer robust måste förbättringar göras. En förbättring är att använda gain scheduling för att anpassa filterkoefficienterna beroende på hastigheten.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-04-12T13:04:06Z No. of bitstreams: 1 alissonborgesdesouza.pdf: 997067 bytes, checksum: 92d74fc8f2a6f1c9ec4f99faf3f94719 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-04-24T03:32:23Z (GMT) No. of bitstreams: 1 alissonborgesdesouza.pdf: 997067 bytes, checksum: 92d74fc8f2a6f1c9ec4f99faf3f94719 (MD5) / Made available in DSpace on 2016-04-24T03:32:23Z (GMT). No. of bitstreams: 1 alissonborgesdesouza.pdf: 997067 bytes, checksum: 92d74fc8f2a6f1c9ec4f99faf3f94719 (MD5) Previous issue date: 2013-03-01 / As proteínas de fontes tradicionais estão sendo cada vez mais utilizadas como ingredientes em um grande número de alimentos formulados. As propriedades tecno-funcionais das proteínas estão relacionadas a várias características moleculares gerais como hidratação, atividade de superfície e tipo de interação proteína-proteína, favorecidas por desdobramento parcial das estruturas. Destas interações o pH do meio apresenta-se como um grande interferente na intensidade e tipo de reação ocorrida. A intensidade de desnaturação e interação das diferentes proteínas ocorre em diferentes formas e intensidade quando se varia o valor de pH do meio em que estas se encontram. Assim, este trabalho teve como objetivo geral a verificação da influência das interações soro-proteínas/caseínas na evolução da viscosidade e perfil de textura de produtos lácteos em diferentes valores de pH, utilizando o simulador o Analisador Rápido de Viscosidade (RVA) e aplicação tecnológica em queijos processados. Foram analisadas amostras comerciais de concentrado protéico de soro (WPC) e concentrado protéico de leite (MPC). A concentração sólido-líquido da solução avaliada no RVA foi 0,3 g.g-1 de matéria seca em água, devido à grande sensibilidade do sistema à concentração de sólidos. A solução teve o pH ajustado para os valores de 6,0, 6,5 e 7,0. Os resultados foram interpretados em função das viscosidades finais após processamento em RVA. Foi possível otimizar as diferentes condições das variáveis dos concentrados protéicos e pH da solução com a interpretação dos perfis viscográficos obtidos, com referência ao comportamento térmico do WPC e do MPC. O RVA pode ser considerado uma ferramenta analítica emergente para o estudo e conhecimento das tecnofuncionalidades, utilizando a análise de viscosidade para estimar a desnaturação protéica. Foi possível também otimizar as diferentes proporções de WPC e MPC para alteração dos parâmetros de textura de queijos processados cremosos, sendo o pH também um fator de influência nesta otimização. / The traditional sources of protein are being increasingly used as ingredients for a large number of formulated foods. The techno-functional properties of proteins are related to various general and molecular characteristics such as hydration, surface activity and type of protein-protein interactions, facilitated by partial unfolding of structures. From these interactions, the medium pH is presented as a major interferer in intensity and type of reaction that takes place. The intensity of denaturation and interaction of different proteins occur in different forms and intensity accordingly to the pH value of the medium in which they are located. This study aimed to verify the influence of interactions between whey /casein in the evolution of the viscosity and texture profile of dairy products at different pH values, using Rapid Viscosity Analyzer (RVA) simulator and its technological application in processed cheese. We have analyzed samples of commercial whey protein concentrate (WPC) and milk protein concentrate (MPC). The solid-liquid concentration of the solution measured in the RVA was 0.3 g.g-1 of dry weight in water due to the high sensitivity of the system to solid concentration. The solution was adjusted to pH values of 6.0, 6.5 and 7.0. The results were interpreted in terms of the final viscosity after processing at RVA. It was possible to optimize the different conditions of variables of concentrated protein solution and of the pH of the solution with the interpretation of the obtained viscosity profiles with reference to the thermal behavior of MPC and WPC. The RVA can be considered an emerging analytical tool for the study and knowledge of the technological features, using the analysis to estimate the viscosity of protein denaturation. It was also possible to optimize the different proportions of WPC and MPC to change the parameters of texture for creamy processed cheese and the pH was also an influencing factor in this optimization.

Ingeniero Civil Electricista / En la actualidad, existe un inmenso interés internacional por el desarrollo de tecnologías de generación con fuentes renovables. Particularmente, la tecnología de concentración solar ha alcanzado suficiente madurez para ser incluida en el mercado de generación y se espera que dentro de los próximos años sea una tecnología presente en la matriz energética de Chile debido principalmente a las condiciones privilegiadas de radiación en el norte del país y también por los incentivos por parte del Estado y también de privados para invertir en esta tecnología. Debido a que en una planta solar no es posible manipular la fuente de energía primaria (radiación solar), el objetivo general del presente trabajo consiste en el diseño de una estrategia de control predictivo basado en modelos (MPC) a nivel supervisor para una planta con colectores cilíndrico parabólicos, que permita aprovechar al máximo el recurso disponible. El trabajo se ha realizado utilizando el simulador de planta solar ACUREX, desarrollado en Matlab-Simulink, el cual representa el comportamiento del lazo de colectores de la plataforma solar de Almería, España. Los modelos predictivos de la planta desarrollados consisten en modelos lineales ARX (autorregresivo con entrada exógena) y ARIX (autorregresivo integral con entrada exógena); y un modelo no lineal difuso de T&S (Takagi & Sugeno), los cuales consideran como variable de salida a la temperatura de salida del aceite del campo colector y como variable de entrada su valor de referencia. Debido a que los modelos ARIX y de T&S presentan los mejores resultados, se propone el diseño de una estrategia de control predictivo a nivel supervisor en base a ambos modelos. Los resultados obtenidos muestran que su implementación reduce el error de seguimiento en 1%, se evita sobreimpulso en el sistema y se limita los cambios en la energía de control, haciendo el funcionamiento del sistema más seguro. En base a pruebas por simulación, se concluye que el MPC difuso supervisor basado en el modelo de T&S mejora el desempeño de la planta frente a cambios en la referencia y en condiciones anormales de operación en comparación con el control regulatorio original. Como líneas de investigación futuras se propone entregar robustez al sistema por medio de la integración de incertidumbre paramétrica en el modelo, incluir criterios económicos de operación e implementar el sistema de control en una futura planta solar en Chile.

Batch and semi-batch processes provide needed flexibility for multi-product plants, especially when products change frequently and production quantities are small. However, challenges occur when trying to implement reliable control systems in batch processes due to some unavoidable inherent characteristics such as the presence of time-varying and nonlinear batch process dynamics and a host of unmeasured disturbances. The most typical control strategy employed in batch process operations does not use utilise online measurements of variables directly related to the product quality and as such is bound to produce off specification products even when the specified control objective has been met. Work done in this thesis is concerned with the design of a supervisory control scheme that takes into consideration the online status of the quality variable of interest from the beginning to the end of the batch process. A novel control methodology is proposed which combines the speed and flexibility of Near-Infrared (NIR) spectroscopic measurements as quality feedback variables within a multiple model predictive control (MPC) framework. In particular the multivariate NIR spectral data is pre-processed for feedback using a statistical model based on Independent Component Analysis (ICA). The proposed controller is tested on a benchmark simulated batch reactor using several case studies and is demonstrated to bring significant improvement in control performance when contrasted with other inferential and direct quality controllers.

Stochastic model predictive control (MPC) is a fascinating field for research and of increasing practical importance since optimal control techniques have been intensively investigated in modern control system design. With the development of computer technologies and communication networks, networked control systems (NCSs) or cyber-physical systems (CPSs) have become an interest of research due to the comprehensive integration of physical systems, such as sensors, actuators and plants, with intricate cyber components, possessing information communication and computation. In CPSs, advantages of low installation cost, high reliability, flexible modularity, improved efficiency, and greater autonomy can be obtained by the tight coordination of physical and cyber components. Several sectors, including robotics, transportation, health care, smart buildings, and smart grid, have witnessed the successful application of CPSs design. The integration of extensive cyber capability and physical plants with ubiquitous uncertainties also introduces concerns over communication efficiency, robustness and stability of the CPSs. Thus, to achieve satisfactory performance metrics of efficiency, robustness and stability, a detailed investigation into control synthesis of CPSs under the stochastic model predictive control framework is of importance. The stochastic model predictive control synthesis plays a vital role in CPSs design since the multivariable stochastic system subject to probabilistic constraints can be controlled in an optimized way. On the other hand, aperiodically sampled, or event-based, model predictive control has also been applied to CPSs extensively to improve communication efficiency. In this thesis, the control synthesis and analysis of aperiodically sampled stochastic model predictive control for CPSs is considered. Chapter 1 provides an introductory literature review of the current development of stochastic MPC, distributed stochastic MPC and event-based MPC. Chapter 2 presents a stochastic self-triggered model predictive control scheme for linear systems with additive uncertainty and with the states and inputs being subject to chance constraints. In the proposed control scheme, the succeeding sampling time instant and current control inputs are computed online by solving a formulated optimization problem. Chapter 3 discusses a stochastic self-triggered model predictive control algorithm with an adaptive prediction horizon. The communication cost is explicitly considered by adding a damping factor in the cost function. Sufficient conditions are provided to guarantee closed-loop chance constraints satisfactions. Furthermore, the recursive feasibility of the algorithm is analyzed, and the closed-loop system is shown to be stable. Chapter 4 proposes a distributed self-triggered stochastic MPC control scheme for CPSs under coupled chance constraints and additive disturbances. Based on the assumptions on stochastic disturbances, both local and coupled probabilistic constraints are transformed into the deterministic form using the tube-based method, and improved terminal constraints are constructed to guarantee the recursive feasibility of the control scheme. Theoretical analysis has shown that the overall closed-loop CPSs are quadratically stable. Numerical examples illustrate the efficacy of the proposed control method in terms of data transmission reductions. Chapter 5 concludes the thesis and suggests some promising directions for future research. / Graduate / 2022-01-15

Traditional robots, while capable of being efficient and effective for the task they were designed, are dangerous when operating in unmodeled environments or around humans. The field of soft robotics attempts to increase the safety of robots thus enabling them to operate in environments where traditional robots should not operate. Because of this, soft robots were developed with different goals in mind than traditional robots and as such the traditional metrics used to evaluate standard robots are not effective for evaluating soft robots. New metrics need to be developed for soft robots so that effective comparison and evaluations can be made. This dissertation attempts to lay the groundwork for that process through a survey on soft robot metrics. Additionally we propose six soft robot actuator metrics that can be used to evaluate and compare characteristics and performance of soft robot actuators. Data from eight different soft robot rotational actuators (five distinct designs) were used to evaluate these soft robot actuator metrics and show their utility. New models, control methods and estimation methods also need to be developed for soft robots. Many of the traditional methods and assumptions for modeling and controlling robotic systems are not able to provide the fidelity that is needed for soft robots to effectively complete useful tasks. This dissertation presents specific developments in each of these areas of soft robot metrics, modeling, control and estimation. We show several incremental improvements to soft robot dynamic models as well as how they were used in control methods for more precise control. We also demonstrate a method for linearizing high degree of freedom models so it can be simplified for use in faster control methods for better performance. Lastly, we present an improved continuum joint configuration estimation method that uses a linear combination of length measurements. All these developments combine to help build the "fundamental engineering framework" that is needed for soft robotics as well as helping to move robots out of their confined spaces and bring them into new unmodeled/unstructured environments.

Collision avoidance is an essential capability for autonomous and assisted-driving ground vehicles. In this work, we developed a novel model predictive control based intelligent collision avoidance (CA) algorithm for a multi-trailer industrial ground vehicle implemented on a General Purpose Graphical Processing Unit (GPGPU). The CA problem is formulated as a multi-objective optimal control problem and solved using a limited look-ahead control scheme in real-time. Through hardware-in-the-loop-simulations and experimental results obtained in this work, we have demonstrated that the proposed algorithm, using NVIDA’s CUDA framework and the NVIDIA Jetson TX2 development platform, is capable of dynamically assisting drivers and maintaining the vehicle a safe distance from the detected obstacles on-thely. We have demonstrated that a GPGPU, paired with an appropriate algorithm, can be the key enabler in relieving the computational burden that is commonly associated with model-based control problems and thus make them suitable for real-time applications.

In this paper, the use of reinforcement learning (RL) in control systems is investigated using a rotatory inverted pendulum as an example. The control behavior of an RL controller is compared to that of traditional LQR and MPC controllers. This is done by evaluating their behavior under optimal conditions, their disturbance behavior, their robustness and their development process. All the investigated controllers are developed using MATLAB and the Simulink simulation environment and later deployed to a real pendulum model powered by a Raspberry Pi. The RL algorithm used is Proximal Policy Optimization (PPO). The LQR controller exhibits an easy development process, an average to good control behavior and average to good robustness. A linear MPC controller could show excellent results under optimal operating conditions. However, when subjected to disturbances or deviations from the equilibrium point, it showed poor performance and sometimes instable behavior. Employing a nonlinear MPC Controller in real time was not possible due to the high computational effort involved. The RL controller exhibits by far the most versatile and robust control behavior. When operated in the simulation environment, it achieved a high control accuracy. When employed in the real system, however, it only shows average accuracy and a significantly greater performance loss compared to the simulation than the traditional controllers. With MATLAB, it is not yet possible to directly post-train the RL controller on the Raspberry Pi, which is an obstacle to the practical application of RL in a prototyping or teaching setting. Nevertheless, RL in general proves to be a flexible and powerful control method, which is well suited for complex or nonlinear systems where traditional controllers struggle.

No description available.

Due to the rapid and prosperous development of information technology, data centers are widely used in every aspect of social life, such as industry, economy or even our daily life. This work considers the idea of developing a data-driven model based model predictive control (MPC) to regulate temperature for a class of single-rack data centers (DCs). An auto-regressive exogenous (ARX) model is identified for our DC system using partial least square (PLS) to predict the behavior of multi-inputs-single-output (MISO) thermal system. Then an MPC controller is designed to control the temperature inside IT rack based on the identified ARX model. Moreover, fuzzy c-means (FCM) is employed to cluster the measured data set. Based on the clustered data sets, PLS is adopted to identify multiple locally linear ARX models which will be combined by appropriate weights in order to capture the nonlinear behavior of the highly-nonlinear thermal system inside the IT rack. The effectiveness of the proposed method is illustrated through experiments on our single-rack DC and it is also compared with proportional-integral (PI) control. / Thesis / Master of Applied Science (MASc)