Este trabalho apresenta técnicas de projeto de controle neural e controle difuso para uma plataforma de Stewart acionada hidraulicamente. O modelo dinâmico não linear da plataforma de Stewart com seis graus de liberdade foi desenvolvido no ambiente de sistemas multicorpos ADAMS. Este pacote comercial foi usado para economizar tempo e esforço na modelagem de um sistema mecânico complexo e na programação para obter a resposta no tempo do sistema. A plataforma de Stewart é um manipulador paralelo com alta relação força-peso e acuracidade de posicionamento comparada a manipuladores seriais convencionais. As desvantagens dos mecanismos seriais é que cada articulação suporta o peso da articulação seguinte e mais o objeto a ser manipulado. A plataforma de Stewart tem recebido recentemente considerável interesse de pesquisadores dado o sucesso de suas aplicações e potencial vantagens sobre os manipuladores convencionais. Uma aplicação bastante popular da plataforma de Stewart é o simulador de vôo onde a plataforma executa movimento com acelerações similares àquelas de uma aeronave. Embora muitas pesquisas na literatura tenham dedicado amplo esforço para cinemática, dinâmica e projeto mecânico de manipuladores baseados em plataforma de Stewart, pouca atenção tem sido dada ao problema de controle deste tipo de manipulador. Um esquema de controle difuso e de redes neurais foi adotado para lidar com as não linearidades, distúrbios e incertezas dos parâmetros, e precisão necessária no posicionamento e orientação da plataforma. Redes neurais artificiais e lógica difusa fornecem um paradigma computacional característico e tem demonstrado resultado para uma faixa de problemas práticos onde a técnica computacional convencional não tem sucesso. Em particular, a habilidade do controle neural e do controle difuso para representar mapeamento não linear encoraja o estudo de controle neural e difuso para representar problemas de controle não linear. Resultados de simulação são apresentados, mostrando que as técnicas propostas podem ser usadas na plataforma de Stewart. / This work presents a neural and fuzzy control design technique for a hydraulically driven Stewart platform. The non-linear dynamic model of a Stewart platform with six degrees of freedom was developed in the multibody systems environment ADAMS. This commercial package was used to save time and effort in modelling the complex mechanical system and in the programming to get the time response of the system. The Stewart platform is a parallel manipulator with high force-to-weight ratio and position accuracy compared to conventional serial manipulators. The disadvantage of serial mechanisms is that each link has to support the weigth of all the following links in addition to the object to be supported. The Stewart platform has recently received considerable research interest due to its successful applications and potential advantages over the conventional manipulators. A quite popular application of the Stewart platform is the flight simulator where the platform performs motion with accelerations similar to those of an airplane. Although much of the research in the literature has devoted extensive effort to the kinematics, dynamics and mechanisms design of the Stewart platform-based manipulators, little attention has been paid to the control problem of this type of manipulators. A fuzzy and neural network control scheme was adopted to deal with the nonlinealities, disturbances and uncertainties of the parameters, and required precision in position and orientation the platform. Artificial neural networks and fuzzy logic provide a distinctive computational paradigm and have proven to be effective for a range of practical problems where conventional computation techniques have not succeeded. In particular, the ability of neural and fuzzy control techniques to represent non-linear mappings encourages the study of these techniques to be used for controling complex non-linear control problems. Simulations results are presented, showing that the proposed technique can be used in a Stewart platform.

<p>Locomotion is one of the most important behaviours and requires interaction between sensors at various levels of the nervous system and the limb muscles of an animal. The basic neural rhythm for locomotion in mammals has been shown to arise from local neural networks residing in the spinal cord and these networks are known as central pattern generators (CPGs). However, during the locomotion, these centres are constantly interacting with the sensory feedback signals coming from muscles, joints and peripheral skin receptors in order to adapt the stepping to varying environmental conditions. Conceptual models of mammalian locomotion have been constructed using</p><p>mathematical models of locomotor subsystems based on the abundance of neurophysiological evidence obtained primarily in the cat. Several aspects of locomotor control using the cat as an animal model have been investigated employing computer simulations and here we use the same approach to address number of questions or/and hypotheses related to rhythmic locomotion in quadrupeds. Some of the involve questions are, role of mechanical linkage during deafferented walking, finding inherent stabilities/instabilities of muscle-joint interactions during normal walking, estimating phase dependent controlability of muscle action over joints.</p><p>This thesis presents the basics of a biologically realistic model of mammalian locomotion and summarises methodological approaches in modelling quadruped locomotor subsystems such as CPGs, limb muscles and sensory pathways. In the first appended article, we extensively discuss the construction details of the three-dimensional computer simulator for the study of the hind leg neuro-musculo-skeletal-control system and its interactions during normal walking of the cat. The simulator with the walking model is programmed in Python scripting language with other supported open source libraries such as Open Dynamics Engine (ODE) for simulating body dynamics and OpenGL for three dimensional graphical representation. We have examined the</p><p>functionality of the simulator and the walking model by simulating deafferented walking. It was possible to obtain a realistic stepping in the hind legs even without sensory feedback to the two controllers (CPGs) for each leg. We conclude that the mechanical linkages between the legs also play a major role in producing alternating gait.</p><p>The use of simulations of walking in the cat for gaining insights into more complex interactions between the environment and the neuro-muscular-skeletal system is important especially for questions where a direct neurophysiological experiment can not be performed on a real walking animal. For instance, it is experimentally hard to isolate individual mechanisms without disrupting the natural walking pattern. In the second article, we introduce a different approach where we use the walking model to identify what control is necessary to maintain stability in the musculo-skeletal system. We show that the actions of most of the hindlimb muscles over the joints have an inherent stability during stepping, even without the involvement of proprioceptive feedback mechanisms. In addition, we observe that muscles generating movements in the ankle joint of the hind leg must be controlled by neural mechanisms, which may involve supraspinal structures, over the whole step cycle.</p>

abstract: The development of advanced, anthropomorphic artificial hands aims to provide upper extremity amputees with improved functionality for activities of daily living. However, many state-of-the-art hands have a large number of degrees of freedom that can be challenging to control in an intuitive manner. Automated grip responses could be built into artificial hands in order to enhance grasp stability and reduce the cognitive burden on the user. To this end, three studies were conducted to understand how human hands respond, passively and actively, to unexpected perturbations of a grasped object along and about different axes relative to the hand. The first study investigated the effect of magnitude, direction, and axis of rotation on precision grip responses to unexpected rotational perturbations of a grasped object. A robust "catch-up response" (a rapid, pulse-like increase in grip force rate previously reported only for translational perturbations) was observed whose strength scaled with the axis of rotation. Using two haptic robots, we then investigated the effects of grip surface friction, axis, and direction of perturbation on precision grip responses for unexpected translational and rotational perturbations for three different hand-centric axes. A robust catch-up response was observed for all axes and directions for both translational and rotational perturbations. Grip surface friction had no effect on the stereotypical catch-up response. Finally, we characterized the passive properties of the precision grip-object system via robot-imposed impulse perturbations. The hand-centric axis associated with the greatest translational stiffness was different than that for rotational stiffness. This work expands our understanding of the passive and active features of precision grip, a hallmark of human dexterous manipulation. Biological insights such as these could be used to enhance the functionality of artificial hands and the quality of life for upper extremity amputees. / Dissertation/Thesis / Ph.D. Mechanical Engineering 2013

Locomotion is one of the most important behaviours and requires interaction between sensors at various levels of the nervous system and the limb muscles of an animal. The basic neural rhythm for locomotion in mammals has been shown to arise from local neural networks residing in the spinal cord and these networks are known as central pattern generators (CPGs). However, during the locomotion, these centres are constantly interacting with the sensory feedback signals coming from muscles, joints and peripheral skin receptors in order to adapt the stepping to varying environmental conditions. Conceptual models of mammalian locomotion have been constructed using mathematical models of locomotor subsystems based on the abundance of neurophysiological evidence obtained primarily in the cat. Several aspects of locomotor control using the cat as an animal model have been investigated employing computer simulations and here we use the same approach to address number of questions or/and hypotheses related to rhythmic locomotion in quadrupeds. Some of the involve questions are, role of mechanical linkage during deafferented walking, finding inherent stabilities/instabilities of muscle-joint interactions during normal walking, estimating phase dependent controlability of muscle action over joints. This thesis presents the basics of a biologically realistic model of mammalian locomotion and summarises methodological approaches in modelling quadruped locomotor subsystems such as CPGs, limb muscles and sensory pathways. In the first appended article, we extensively discuss the construction details of the three-dimensional computer simulator for the study of the hind leg neuro-musculo-skeletal-control system and its interactions during normal walking of the cat. The simulator with the walking model is programmed in Python scripting language with other supported open source libraries such as Open Dynamics Engine (ODE) for simulating body dynamics and OpenGL for three dimensional graphical representation. We have examined the functionality of the simulator and the walking model by simulating deafferented walking. It was possible to obtain a realistic stepping in the hind legs even without sensory feedback to the two controllers (CPGs) for each leg. We conclude that the mechanical linkages between the legs also play a major role in producing alternating gait. The use of simulations of walking in the cat for gaining insights into more complex interactions between the environment and the neuro-muscular-skeletal system is important especially for questions where a direct neurophysiological experiment can not be performed on a real walking animal. For instance, it is experimentally hard to isolate individual mechanisms without disrupting the natural walking pattern. In the second article, we introduce a different approach where we use the walking model to identify what control is necessary to maintain stability in the musculo-skeletal system. We show that the actions of most of the hindlimb muscles over the joints have an inherent stability during stepping, even without the involvement of proprioceptive feedback mechanisms. In addition, we observe that muscles generating movements in the ankle joint of the hind leg must be controlled by neural mechanisms, which may involve supraspinal structures, over the whole step cycle. / QC 20101111

Electrical activity of the stomach is one determining factor of gastric motility by controlling and coordinating contractions of the gastric musculature. These contractions, both tonic and phasic, are responsible for the storing, mixing, and emptying of food. Gastric electrical activity is therefore a very important factor for normal stomach function. The development of a multi-channel, bandlimited, signal amplifier and recording system, provides a means to record this electrical activity. Many practical issues are addressed to provide a signal of acceptable quality and several basic signal processing techniques are applied to increase the quality of these signals and provide extraction of important information regarding power and frequency content. Gastric electrical activity is recorded from the stomachs of several rats in various experiments. The recorded activity in different regions of the stomach, responsible for different functions, is compared and evaluated with respect to known cellular events. By introducing several stimuli and observing changes in recorded activity, the nervous control of the stomach via mediation of the electrical activity is also examined and modeled briefly. / Thesis / Master of Engineering (ME)

Made available in DSpace on 2014-12-17T14:55:03Z (GMT). No. of bitstreams: 1 OscarGF.pdf: 1901439 bytes, checksum: f8f1a37dca7a69d726f7a9453cbf0a98 (MD5) Previous issue date: 2004-03-05 / This work develops a robustness analysis with respect to the modeling errors, being applied to the strategies of indirect control using Artificial Neural Networks - ANN s, belong to the multilayer feedforward perceptron class with on-line training based on gradient method (backpropagation). The presented schemes are called Indirect Hybrid Control and Indirect Neural Control. They are presented two Robustness Theorems, being one for each proposed indirect control scheme, which allow the computation of the maximum steady-state control error that will occur due to the modeling error what is caused by the neural identifier, either for the closed loop configuration having a conventional controller - Indirect Hybrid Control, or for the closed loop configuration having a neural controller - Indirect Neural Control. Considering that the robustness analysis is restrict only to the steady-state plant behavior, this work also includes a stability analysis transcription that is suitable for multilayer perceptron class of ANN s trained with backpropagation algorithm, to assure the convergence and stability of the used neural systems. By other side, the boundness of the initial transient behavior is assured by the assumption that the plant is BIBO (Bounded Input, Bounded Output) stable. The Robustness Theorems were tested on the proposed indirect control strategies, while applied to regulation control of simulated examples using nonlinear plants, and its results are presented / Este trabalho utiliza as Redes Neurais Multicamadas - RNM s, totalmente com treinamento em tempo real (on-line), no desenvolvimento de duas estrat?gias de controle indireto. Os esquemas propostos denominam-se Controle H?brido Indireto e Controle Neural Indireto. Todo o treinamento dos neurodispositivos - o identificador da planta e o controlador, quando presentes na malha de controle indireto, ? realizado com um m?nimo de atraso computacional, de modo a contemplar o controle de plantas com pequenos per?odos de amostragem. S?o apresentados Teoremas de Estabilidade para garantia da converg?ncia dos dispositivos neurais, assim como foram feitas considera??es para adequar o m?todo de acelera??o da converg?ncia h-adaptativo utilizado ?s condi??es de estabilidade. Para cada esquema de controle indireto foi desenvolvido um teorema que permite calcular o m?ximo erro permanente (steady-state error) que poder? ocorrer em fun??o da toler?ncia previamente especificada para converg?ncia dos dispositivos neurais usados na malha de controle, desde que a estabilidade seja garantida. Estes teoremas foram denominados de Teoremas da Robustez e constituem a principal contribui??o deste trabalho. As condi??es de estabilidade e robustez foram testadas para as estrat?gias de Controle H?brido Indireto e de Controle Neural Indireto, sendo apresentados os resultados obtidos na simula??o computacional do controle de regula??o de plantas n?o-lineares, BIBO (Bounded Input, Bounded Output) est?veis

The main goal of the dissertation was to determine task-dependent modulation of corticospinal descending output. From this main goal, I conducted three different studies to determine how corticospinal output to muscles of the upper arm and hand changed as a function of the task demands. In study 1, I examined how a somatosensory-motor circuit changes when a muscle needs to be active in a task and found that this circuit may be dependent on the movement phase, type of afferent input, and the task demands. In study 2, I examined how this same somatosensory-motor circuit acts to both allow and prevent muscle activity before movement. I revealed that this somatosensory-motor circuit may function to prevent muscle activity when a muscle is not needed in a task and creates facilitation of corticospinal output when it needs to be active in a task. These effects, however, are dependent on the movement phase and the digit the muscle is controlling. Study 3 determined how corticospinal output is modulated to upper arm muscles when performing movements that required different combinations of segmental interactions to achieve the task successfully. Corticospinal output was increased when inertia and the BBC moment at a joint resisted the intended joint rotation and these effects were dependent on the muscle and movement phase. I propose a model of the connectivity between the primary motor and somatosensory cortices that would increase, modulate, or decrease corticospinal output to a muscle depending on its role in the task. The findings from this work provides information to guide future neural rehabilitative interventions for individuals who have movement disorders arising from altered somatosensory-motor processing such as Cerebellar Ataxia, Developmental Coordination Disorder, Focal Hand Dystonia, Parkinson’s disease, and stroke. / Dissertation / Doctor of Philosophy (PhD) / On a day to day basis, we perform a variety of movements without giving much thought to how complicated it is for our nervous system to perform said movements. There are many different areas of the brain that are responsible for controlling movement. This dissertation focuses on two key areas that are critical for movement performance, namely the primary motor and somatosensory cortices. The primary motor cortex is largely responsible for sending signals to the muscles to control movement, while the primary somatosensory cortex plays a crucial role in receiving and understanding sensory input from our body. The studies in this dissertation describe how these two areas of the brain communicate during finger and arm movements to produce or prevent muscle activity. This work has implications for individuals with disorders that impact their everyday movements.

Optogenetics is a set of technologies that enable optically triggered gain or loss of function in genetically specified populations of cells. Optogenetic methods have revolutionized experimental neuroscience by allowing precise excitation or inhibition of firing in specified neuronal populations embedded within complex, heterogeneous tissue. Although optogenetic tools have greatly improved our ability manipulate neural activity, they do not offer control of neural firing in the face of ongoing changes in network activity, plasticity, or sensory input. In this thesis, I develop a feedback control technology that automatically adjusts optical stimulation in real-time to precisely control network activity levels. I describe hardware and software tools, modes of optogenetic stimulation, and control algorithms required to achieve robust neural control over timescales ranging from seconds to days. I then demonstrate the scientific utility of these technologies in several experimental contexts. First, I investigate the role of connectivity in shaping the network encoding process using continuously-varying optical stimulation. I show that synaptic connectivity linearizes the neuronal response, verifying previous theoretical predictions. Next, I use long-term optogenetic feedback control to show that reductions in excitatory neurotransmission directly trigger homeostatic increases in synaptic strength. This result opposes a large body of literature on the subject and has significant implications for memory formation and maintenance. The technology presented in this thesis greatly enhances the precision with which optical stimulation can control neural activity, and allows causally related variables within neural circuits to be studied independently.

Made available in DSpace on 2014-12-17T14:55:56Z (GMT). No. of bitstreams: 1 OdailsonCO_DISSERT.pdf: 807518 bytes, checksum: 6b1541fcc8435e2f642b214f2496487c (MD5) Previous issue date: 2011-07-28 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / An alternative nonlinear technique for decoupling and control is presented. This technique is based on a RBF (Radial Basis Functions) neural network and it is applied to the synchronous generator model. The synchronous generator is a coupled system, in other words, a change at one input variable of the system, changes more than one output. The RBF network will perform the decoupling, separating the control of the following outputs variables: the load angle and flux linkage in the field winding. This technique does not require knowledge of the system parameters and, due the nature of radial basis functions, it shows itself stable to parametric uncertainties, disturbances and simpler when it is applied in control. The RBF decoupler is designed in this work for decouple a nonlinear MIMO system with two inputs and two outputs. The weights between hidden and output layer are modified online, using an adaptive law in real time. The adaptive law is developed by Lyapunov s Method. A decoupling adaptive controller uses the errors between system outputs and model outputs, and filtered outputs of the system to produce control signals. The RBF network forces each outputs of generator to behave like reference model. When the RBF approaches adequately control signals, the system decoupling is achieved. A mathematical proof and analysis are showed. Simulations are presented to show the performance and robustness of the RBF network / Neste trabalho, ser? apresentada uma t?cnica alternativa para o desacoplamento e controle de sistemas n?o lineares. A estrat?gia de desacoplamento proposta est? baseada numa rede neural RBF (Radial Basis Functions) combinada com o controle adaptativo por modelo de refer?ncia. A t?cnica ? aplicada no controle do modelo de um gerador s?ncrono, cujas vari?veis de sa?da s?o o ?ngulo de carga e o fluxo concatenado no enrolamento de campo. O sistema do gerador s?ncrono ? acoplado, ou seja, a mudan?a numa das vari?veis de entrada do sistema altera mais de uma vari?vel de sa?da. A rede RBF realizar? o desacoplamento do sistema, fazendo o controle de forma independente de cada uma das sa?das. Tal estrat?gia n?o exige conhecimento dos par?metros do sistema e observa-se um comportamento est?vel da rede RBF, tanto na presen?a de incertezas na modelagem, como de perturba??es no sistema. Ser? mostrada a simplicidade da aplica??o da t?cnica e do projeto da rede RBF. Os pesos, que interligam as camadas oculta e de sa?da da rede, s?o ajustados utilizando uma lei adaptativa em tempo real. Essa lei adaptativa foi desenvolvida pelo m?todo de fun??es de energia de Lyapunov. O sistema de controle e desacoplamento faz uso dos sinais filtrados da sa?da do gerador e dos sinais dos erros entre as sa?das do gerador e as sa?das do modelo refer?ncia. Assim, atrav?s dos sinais de controle aplicados pela rede RBF, cada sa?da do sistema do gerador ? for?ada a se comportar conforme uma din?mica desejada, dada pelo modelo de refer?ncia. Quando a rede RBF aproxima adequadamente os sinais de controle, o desacoplamento do sistema ? alcan?ado. Os resultados do desempenho da estrat?gia ser?o apresentados atrav?s de simula??es. Tamb?m ser? mostrada a prova matem?tica de estabilidade do sistema em malha fechada para o caso escalar

No description available.