431 |
[en] AUV AUTO-DOCKING APPROACH BASED ON REINFORCEMENT LEARNING AND VISUAL SERVOING / [pt] TÉCNICA DE ACOPLAGEM AUTOMÁTICA DE AUV BASEADA EM APRENDIZADO POR REFORÇO E SERVOVISÃOMATHEUS DO NASCIMENTO SANTOS 24 January 2024 (has links)
[pt] No campo em crescimento da robótica subaquática, Veículos Subaquáticos
Automatizados (AUVs) estão se tornando cada vez mais importantes para
uma variedade de usos, como exploração, mapeamento e inspeção. Esta dissertação
foca em estudar os principais desafios da acoplagem automática de AUVs,
considerando um ambiente 3D simulado personalizado. A pesquisa divide essa
tarefa em duas partes principais: estimativa da pose da garagem e estratégia
de controle do AUV. Utilizando uma mistura de métodos tradicionais e novos,
incluindo sistemas baseados em marcos fiduciais, Redes Neurais Convolucionais
(CNN) e Aprendizado por Reforço (RL), o estudo realiza experimentos
para verificar o desempenho e as limitações do sistema.
Um aspecto significativo desta dissertação é o uso de um ambiente 3D
simulado para facilitar o desenvolvimento e o teste de algoritmos de acoplagem
automática para AUVs. Este ambiente simula dinâmicas subaquáticas,
sensores robóticos e atuadores, permitindo experimentar diferentes técnicas de
estimativa de pose e estratégias de controle. Além disso, o estabelecimento
de um ambiente 3D simulado amigável para RL representa uma contribuição
relevante, oferecendo uma plataforma reutilizável que não apenas valida os algoritmos
de acoplagem automática desenvolvidos neste estudo, mas também
serve como base para futuras aplicações subaquáticas baseadas em RL.
Em resumo, a dissertação explora uma série de cenários para avaliar a
eficácia de várias técnicas de acoplagem automática. Inicialmente, ela utiliza
servo-visualização junto com um controlador PID tradicional, seguido pela
introdução de métodos mais avançados, como estimadores de pose baseados
em CNN e controladores de Aprendizado por Reforço. Esses métodos são
avaliados tanto individualmente quanto em combinações híbridas para medir
sua adequação e limitações para entender os principais desafios por trás da
acoplagem automática de AUVs. / [en] In the growing field of underwater robotics, Automated Underwater
Vehicles (AUVs) are becoming more important for a range of uses, such as
exploration, mapping, and inspection. This dissertation focuses on studying
the main challenges of AUV auto-docking, considering a customized 3D
simulated environment. The research breaks down this challenging task into
two main parts: cage pose estimation and AUV control strategy. Using a mix of
traditional and new methods, including fiducial-based systems, Convolutional
Neural Networks (CNN), and Reinforcement Learning (RL), the study carries
out experiments to check system performance and limitations.
A significant aspect of this dissertation is using a 3D simulated environment
to facilitate the development and testing of auto-docking algorithms
for AUVs. This environment simulates crucial underwater dynamics, robotic
sensors, and actuators, allowing for experimenting with different pose estimation
techniques and control strategies. Additionally, the establishment of an
RL-friendly 3D simulated environment stands as a relevant contribution, offering
a reusable platform that not only validates the auto-docking algorithms
developed in this study but also serves as a foundation for future RL-based
underwater applications.
In summary, the dissertation explores a range of scenarios to evaluate the
efficacy of various auto-docking techniques. It initially utilizes visual servoing
along with a traditional PID controller, followed by the introduction of more
advanced methods like CNN-based pose estimators and Reinforcement Learning
controllers. These methods are assessed both individually and in hybrid
combinations to gauge their suitability and limitations for understanding the
main challenges behind the AUV auto-docking.
|
432 |
Veratridine Can Bind to a Site at the Mouth of the Channel Pore at Human Cardiac Sodium Channel NaV1.5Gulsevin, Alican, Glazer, Andrew M., Shields, Tiffany, Kroncke, Brett M., Roden, Dan M., Meiler, Jens 20 January 2024 (has links)
The cardiac sodium ion channel (NaV1.5) is a protein with four domains (DI-DIV), each
with six transmembrane segments. Its opening and subsequent inactivation results in the brief rapid
influx of Na+ ions resulting in the depolarization of cardiomyocytes. The neurotoxin veratridine
(VTD) inhibits NaV1.5 inactivation resulting in longer channel opening times, and potentially fatal
action potential prolongation. VTD is predicted to bind at the channel pore, but alternative binding
sites have not been ruled out. To determine the binding site of VTD on NaV1.5, we perform docking
calculations and high-throughput electrophysiology experiments in the present study. The docking
calculations identified two distinct binding regions. The first site was in the pore, close to the
binding site of NaV1.4 and NaV1.5 blocking drugs in experimental structures. The second site was at
the “mouth” of the pore at the cytosolic side, partly solvent-exposed. Mutations at this site (L409,
E417, and I1466) had large effects on VTD binding, while residues deeper in the pore had no effect,
consistent with VTD binding at the mouth site. Overall, our results suggest a VTD binding site
close to the cytoplasmic mouth of the channel pore. Binding at this alternative site might indicate an
allosteric inactivation mechanism for VTD at NaV1.5
|
433 |
Investigation of Protein/Ligand Interactions Relating Structural Dynamics to Function: Combined Computational and Experimental ApproachesPavlovicz, Ryan Elliott 24 June 2014 (has links)
No description available.
|
434 |
Identification of novel monoamine oxidase B inhibitors from ligand based virtual screeningAlaasam, Mohammed 30 July 2014 (has links)
No description available.
|
435 |
Inhibition of monoamine oxidase by derivatives of piperine, an alkaloid from the pepper plant Piper nigrum, for possible use in Parkinson’s diseaseAl-Baghdadi, Osamah Basim Khalaf 27 October 2014 (has links)
No description available.
|
436 |
A Computational Investigation Into the Development of an Effective Therapeutic Against Organophosphorus Nerve Agent ExposureBrown, Jason David January 2014 (has links)
No description available.
|
437 |
Modeling and Analysis of Ligand Docking to Norovirus Capsid Protein for the Computer-Aided Drug DesignCHHABRA, MONICA 28 August 2008 (has links)
No description available.
|
438 |
Computational And Experimental Studies Towards The Development Of Novel Therapeutics Against Organophosphorus Nerve Agents: Butyrylcholinesterase And ParaoxonaseVyas, Shubham 12 September 2011 (has links)
No description available.
|
439 |
Multiple Ligand Simultaneous Docking (MLSD) and Its Applications to Fragment Based Drug Design and Drug RepositioningLi, Huameng 06 January 2012 (has links)
No description available.
|
440 |
Autonomous Docking of Electric Boat / Autonom tilläggning av elektrisk båtBOCZAR, LUDVIG, PERNOW, JONATHAN January 2021 (has links)
In recreational boating, docking is one of the most stressful and accident prone situations. Due to the loss of maneuverability at low speeds, it is a procedure that requires experience. There are mainly two problems when it comes to autonomous docking of a boat, these are identifying a berth’s position as well as keeping the boat on its intended path and correcting any deviations. Autonomous docking in recreational boating is still quite uncommon, with companies still exploring different solutions. This thesis proposes a Model Predictive Control (MPC) system combined with Pulsed Coherent Radar technology, equipped on an under-actuated boat model, to achieve autonomous docking. A major part of this thesis was to evaluate the amount and placement of radar sensors, as well as determining whether these are suitable in a water environment. In order to test this, the sensors were placed alongside the hull of the boat. It was found that the placement of sensors had a bigger impact than the amount when it came to correctly detecting the position of a berth. Once the placement of sensors and the berth position algorithmhad been done, a closed-loop MPC was used. This controller got constant feedback of the boat’s position relative the berth, in order to calculate the thruster control inputs for the next time step. The developed autonomous docking system was then implemented on the boat which was tested in a swimming pool. The optimal radar configuration combined withMPC, made it possible to successfully dock a boat autonomously without any modification to the berth. / För fritidsbåtlivet är tilläggning en av demest stressfulla och olycksbenägna situationerna. På grund av förlust av manövrering vid låga hastigheter är det en procedur som kräver erfarenhet. Det finns främst två problem när det kommer till autonom tilläggning, det är att identifiera positionen av en brygga såväl som att hålla båten på den avsedda kursen och rätta till små avvikelser. Autonom tilläggning för fritidsbåtlivet är fortfarande rätt ovanligt och företag utforskar fortfarande olika lösningar. Denna avhandling föreslår ett Modellprediktivt Reglersystem (MPC) kombinerat med Pulserad Koherent Radarteknik som är utrustad på en underaktuerad båtmodell för att uppnå autonom tilläggning. En stor del av avhandlingen var att utvärdera antalet och placeringen av radarsensorer, såväl som att fastställa om dessa är lämpliga att användas i en vattenmiljö. För att undersöka detta placerades sensorerna längs med båtens skrov. Det konstaterades att placeringen av sensorer hade en större påverkan än mängden när det kom till att läsa av positionen av bryggan korrekt. När placeringen av sensorer och bryggpositionsalgoritmen var klar användes MPC med återkoppling. Denna regulator fick konstant återkoppling av båtens position relativt bryggan för att räkna ut styrsignal till motorerna för nästa tidssteg. Den utvecklade autonoma tilläggningen var sedan implementerad på båten som testades i en pool. Den optimala radarplaceringen kombinerat med MPC gjorde det möjligt att med framgång kunna lägga till båten autonomt utan modifiering av bryggan.
|
Page generated in 0.3923 seconds