Thesis (M.S.)--University of Missouri--Rolla, 2007. / Vita. The entire thesis text is included in file. Title from title screen of thesis/dissertation PDF file (viewed February 4, 2008) Includes bibliographical references.

In this thesis, we present research on synthesizing autonomous driving with shared autonomy using Unity Engine. Adaptive Cruise Control (ACC) is considered as level 1 autonomous vehicle, which has been studied by academia and commercialized by industry. Cooperative Adaptive Cruise Control (CACC) system is an expansion of ACC, in which communication is set up between members to share driving information. Shared autonomy is a subject about human-computer interactivities. In our research, we developed a highly customizable 3D environment. We can simulate various driving scenarios and analyze the performance of different driving methods from human driving to CACC. The result of simulation proves the safety and efficiency of CACC, and the project also provides a potential of assisting the improvement of autonomous vehicles.

No description available.

Apresenta uma solução inovadora de controle por redes neurais artificiais aprendendo segundo a técnica de aprendizagem por reforço usando algoritmos genéticos para integrar múltiplos sistemas ativos no controle de estabilidade de um veículo. Estudo, restringido a um domínio de manobras, foi desenvolvido excluindo falhas e alterações da planta no tempo. Contribui para responder como o controlador de dinâmica veicular pode ser aperfeiçoado para atuação simultânea de múltiplos sistemas ativos. Contempla o desenvolvimento do neurocontrolador e algoritmo de aprendizagem na plataforma Matlab, de um modelo de dinâmica veicular em ambiente ADAMS e do modelo de referência, atuadores e observador com programação Matlab. Analisa a estabilidade da planta e define regiões de atuação do controlador. Apresenta um estudo e definição da técnica de controle de estabilidade em guinada para nortear a função de otimização, o treinamento e as simulações. Treinamento da rede neural para acomodar as não linearidades envolvidas na planta e para otimizar a integração dos múltiplos sistemas ativos focando nas especificações de desempenho do controlador e no domínio de situações a serem analisadas. Simulação de situações e manobras para validação e avaliação do desempenho do controlador com co-simulação entre Matlab e ADAMS. Resultados qualitativos e quantitativos do desempenho do controlador justificando a integração efetiva dos sistemas e o neurocontrolador não-linear. / Presents an innovative control solution with artificial neural networks learning using reinforcement learning by genetic algorithms to integrate multiple active systems to control yaw vehicle stability. Study restricted to a maneuver domain and excluding plant changes in time and failures. Contributes to answer how the vehicle dynamics controller can be improved for multiple simultaneous active systems. Development of the neurocontroller and learning algorithm in Matlab, vehicle dynamics model in ADAMS environment and reference model, actuators and observer with Matlab programming. Plant stability analysis and activation areas definition. Study and method definition for stability yaw control to guide the task of optimization, training and simulation. Training the neural network to accomplish the plant nonlinearity and to optimize the multiple active systems synergy targeting the controller performance specifications and the analyzed conditions domain. Conditions and maneuvers simulation to validate and evaluate the controller performance using cosimulation between Matlab and ADAMS. Qualitative and quantitative controller results justifying the effective systems integration and non-linear neurocontroller.

The purpose of the thesis is to achieve a hybrid force and motion control method of parallel manipulators working in a constrained environment, in the presence of joint flexibility that occurs at the actuated joints. A flexible joint is modeled and the equations of motion of the parallel manipulator are derived by using the Lagrange formulation. The structural damping of the active joints, viscous friction at the passive joints and the rotor damping are also considered in the model. It is shown that in a flexible joint manipulator, the acceleration level inverse dynamic equations are singular because the control torques do not have instantaneous effect on the manipulator end-effector contact forces and accelerations due to the flexibility. Implicit numerical integration methods are utilized for solving the singular equations. As a case study, a two legged constrained planar parallel manipulator with three degrees of freedom is simulated to illustrate the performance of the method.

Made available in DSpace on 2014-06-11T19:27:14Z (GMT). No. of bitstreams: 0 Previous issue date: 2004-05-28Bitstream added on 2014-06-13T20:16:17Z : No. of bitstreams: 1 grassi_lfm_me_ilha.pdf: 1030941 bytes, checksum: 1b6f71060acf49c7dfe0879c783856af (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A necessidade de controle no tratamento de sistemas dinâmicos, com complexidade crescente e diante de fatores de incerteza, tem levado à reavaliação dos métodos convencionais e à proposição de métodos conceitualmente mais elaborados de controle. Estas novas propostas incluem, por exemplo, níveis hierárquicos de decisão, planejamento e aprendizagem, que são necessários quando um alto grau de autonomia do sistema é desejável. Assim as metodologias baseadas em redes neurais, que utilizam modelos matemáticos e técnicas numéricas inspiradas no cérebro humano e/ou sistema nervoso, representam um passo natural na evolução da teoria de controle, principalmente junto àqueles que envolvem não-linearidades. Este trabalho apresenta um estudo da técnica denominada wavenet, que combina redes neurais e transformada wavelet, como um direcionamento alternativo para a solução de problemas de identificação e controle de plantas não lineares. A transformada wavelet utiliza janelas com escala variável que possibilitam analisar faixas de altas e baixas freqüências em um mesmo sinal, e é exatamente essa capacidade de manipulação da janela de observação que a torna uma boa alternativa como função de ativação, realizando um mapeamento local do sinal. Isso proporciona uma identificação mais eficiente, principalmente em sinais não lineares e variantes no tempo. Vários testes simulados envolvendo não linearidade foram analisados visando estudar o comportamento do algoritmo wavenet e definir quais os tipos de funções de ativação, Morlet, Rasp ou Polywog, poderiam fornecer melhores resultados. Utilizou-se o método de otimização de Levenberg-Marquadt, o qual apresentou um desempenho melhor quando comparado com o método do gradiente descendente utilizado por outros autores, no processo de minimização do erro entre a saída da rede e a... . / The necessity of dynamic systems treatment control, with upper complexity and uncertain factors, has lead to reevaluation of conventional methods and the proposition of conceptly methods more elaborate of control. These new proposals include, for instance, hierarchic levels of decision, planning and learning, which are needed when a high degree of system autonomy is desirable. Thus the methodologies based in neural nets, which use mathematical models and numerical techniques inspired in human brain and/or nervous system, represent a natural step in evolution of control theory, mainly join to those which involve no-linearity. This work shows a technique study called wavenet, it combine neural nets and wavelet transformed, as an alternative leading for the solution of identification problems and non linear plants control. The transformed wavelet uses windows with variable scale and it makes possible analyze strips high and low frequencies in a same signal, and it is exactly this capacity of manipulation of observation window and it becomes a good alternative as activation function, achieving a local map of the signal. A identification more efficient is provided, mainly in non-linear signals and time variants. Several simulate tests involving non linear was analyzed, seeking to study the behavior of the algorithm wavenet and to define which the types of activation functions, Morlet, Rasp or Polywog, could give better results. The optimization method of Levenberg-Marquadt was used, and that one show a better performance when compared with the descendent gradient method used by other authors, in minimization of error process between the net and plant exit. The tests looked for to define improvements in algorithm wavenet, in relation to identification process, because it is primordial stage in the project of neurocontrolers. The... (Complete abstract, click electronic address below).

Apresenta uma solução inovadora de controle por redes neurais artificiais aprendendo segundo a técnica de aprendizagem por reforço usando algoritmos genéticos para integrar múltiplos sistemas ativos no controle de estabilidade de um veículo. Estudo, restringido a um domínio de manobras, foi desenvolvido excluindo falhas e alterações da planta no tempo. Contribui para responder como o controlador de dinâmica veicular pode ser aperfeiçoado para atuação simultânea de múltiplos sistemas ativos. Contempla o desenvolvimento do neurocontrolador e algoritmo de aprendizagem na plataforma Matlab, de um modelo de dinâmica veicular em ambiente ADAMS e do modelo de referência, atuadores e observador com programação Matlab. Analisa a estabilidade da planta e define regiões de atuação do controlador. Apresenta um estudo e definição da técnica de controle de estabilidade em guinada para nortear a função de otimização, o treinamento e as simulações. Treinamento da rede neural para acomodar as não linearidades envolvidas na planta e para otimizar a integração dos múltiplos sistemas ativos focando nas especificações de desempenho do controlador e no domínio de situações a serem analisadas. Simulação de situações e manobras para validação e avaliação do desempenho do controlador com co-simulação entre Matlab e ADAMS. Resultados qualitativos e quantitativos do desempenho do controlador justificando a integração efetiva dos sistemas e o neurocontrolador não-linear. / Presents an innovative control solution with artificial neural networks learning using reinforcement learning by genetic algorithms to integrate multiple active systems to control yaw vehicle stability. Study restricted to a maneuver domain and excluding plant changes in time and failures. Contributes to answer how the vehicle dynamics controller can be improved for multiple simultaneous active systems. Development of the neurocontroller and learning algorithm in Matlab, vehicle dynamics model in ADAMS environment and reference model, actuators and observer with Matlab programming. Plant stability analysis and activation areas definition. Study and method definition for stability yaw control to guide the task of optimization, training and simulation. Training the neural network to accomplish the plant nonlinearity and to optimize the multiple active systems synergy targeting the controller performance specifications and the analyzed conditions domain. Conditions and maneuvers simulation to validate and evaluate the controller performance using cosimulation between Matlab and ADAMS. Qualitative and quantitative controller results justifying the effective systems integration and non-linear neurocontroller.

Electric grids across the world are experiencing an ever increasing number of extreme events ranging from extreme weather events to cyberattacks. Such extreme events have the potential to cause widespread power outages and even a blackout. A vast majority of power outages impacting the U.S. electric grid impact the distribution system. There are an estimated five million miles of distribution lines in the US electric grid. A majority of these lines are low-clearance overhead lines making them even more susceptible to damage during extreme events. However, this vital component of the U.S. electric grid remained neglected until recently. In recent decades, the integration of distributed energy resources (DERs) such as solar photovoltaic systems and battery energy storage systems at the grid edge have provided a major opportunity for enhancing the resilience of distribution systems. These DERs can be used to restore power supply when the bulk grid becomes unavailable. However, managing the interactions among different types of DERs has been challenging. Low inertia and significant differences in time constants of operation between conventional generation and inverter based resources (IBRs) are some of these challenges. Widespread deployment of microgrid controller capabilities can be a promising solution to manage these interactions. However, due to interoperability and integration challenges of optimization and dynamics control systems, power conversion systems and communication systems, the adoption of microgrids especially in underserved communities has been slow. The research presented in this dissertation is a significant step forward in this direction by proposing an approach which integrates optimal dispatch, sequential microgrid restoration and control algorithms. Potential cyberattack paths are identified by creating a detailed cyber-physical system model for microgrids. A two-tiered intrusion detection system is developed to detect and mitigate cyberattacks within the cyber layer itself. The developed sequential microgrid restoration algorithm coordinates optimal DER dispatch with the operation of legacy devices with no remote control or communication capabilities and net-metered loads with limited communications. By better utilizing the control capabilities of IBRs, reliance on low-latency centralized control algorithms has also been reduced. The developed approach systematically ensures adequate availability of control during dispatch and restoration to maintain microgrid stability. This research can thus pave the way for faster and more cost-effective deployment of microgrids. / Doctor of Philosophy / A U.S. National Academy of Engineering report has described the power grid as the greatest engineering achievement of the 20th century. The power grid is a complex interconnected system consisting of the power transmission system and the distribution system. The power transmission system consists of the power lines seen while driving on the freeways and the large power generating stations consisting of renewable, coal or nuclear power plants. Ensuring the reliable operation of the transmission system has always been a priority. The distribution system on the other hand consists of pole top transformers seen closer to homes which reduce the voltage to levels safe for electrical appliances. It also consists of the millions of miles of low-clearance overhead distribution lines deployed across the U.S. that provide electricity to every household. This critical part of U.S. electricity infrastructure had remained neglected which is the reason why 90% of power outages impact the distribution system. In recent decades, the integration of renewable energy sources like solar systems and battery storage systems has created an unprecedented opportunity for increasing the resilience of distribution systems against extreme events. These energy sources can provide power supply when the transmission system becomes unavailable. However, ensuring safe and reliable integrated operation of these sources with conventional diesel generators especially while isolated from the transmission system is challenging. This is where microgrids, which are self-sufficient miniature power grids, can help. Microgrids provide required control, communication and cybersecurity features necessary for reliable integrated operation of renewable and conventional energy sources. However, the challenges involved with interoperability of these systems has slowed down the deployment of microgrids especially in underserved communities. This is the research gap addressed in this dissertation. This research provides an approach for integrating the optimization, control, power electronics and cybersecurity systems. Reliance on expensive low-latency communication systems is reduced by utilizing the emerging capabilities of power electronics devices used for integrating the renewable energy sources with the electric power grid. Voltage control devices already deployed in the distribution systems which do not have remote control or communication capabilities have also been coordinated with energy sources. The research presented in this dissertation is a significant step forward for increasing access to power supply during outages and for reducing the time and cost of deployment of microgrids.

The purpose of this thesis is to develop a position control method for parallel manipulators so that the end effector can follow a desired trajectory specified in the task space where joint flexibility that occurs at the actuated joints is also taken into consideration. At the beginning of the study, a flexible joint is modeled, and the equations of motion of the parallel manipulators are derived for both actuator variables and joint variables by using the Lagrange formulation under three assumptions regarding dynamic coupling between the links and the actuators. These equations of motion are transformed to an input/output relation between the actuator torques and the actuated joint variables to achieve the trajectory tracking control. Moreover, the singular configurations of the parallel manipulators are explained. As a case study, a three degree of freedom, two legged planar parallel manipulator is simulated considering joint flexibility. The structural damping of the active joints, viscous friction at the passive joints and the rotor damping are also considered throughout the study. Matlab&reg / and Simulink&reg / softwares are used for the simulations. The results of the simulations reveal that steady state errors are negligibly small and good tracking performances can be achieved.

In this thesis, a technique for the motion of parallel manipulators through drive singularities is investigated. To remedy the problem of unbounded inverse dynamics solution in the neighborhood of drive singularities, an inverse dynamics controller which uses a conventional inverse dynamics control law outside the neighborhood of singularities and switches to the mode based on the formerly derived modified equations inside the neighborhood of singularities is proposed. As a result, good tracking performance is obtained while the actuator forces remain within the saturation limits of the actuators around singular configurations.