Download or read book Partially Observable Markov Decision Processes for Fault Management in Autonomous Underwater Vehicles written by Kathleen Svendsen and published by . This book was released on 2019 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: Proposed, is a partially observable Markov decision process (POMDP) model-based schema as the basis of a fault manager system for use by autonomous underwater vehicles (AUV) undergoing long endurance missions with the operator far from the AUV. The thesis explains the reasoning behind using POMDP over traditional static look-up tables at achieve a more autonomous system. The objective was to develop POMDP models for two illustrative AUV sub-systems - depth and power-management. These models were used as fault managers for a series of simulations for each sub-system, individually, and then when there are interactions. This novel solution demonstrated the validity of POMDP as the basis for a fault manager in accounting for the inherent partial observability of AUV states and their environments. Future work aims to expand this with more AUV sub-systems and test on hardware-in-the-loop simulators.

Download or read book Sensor Resource Management Via Partially Observable Markov Decision Processes written by Lucas W. Krakow and published by . This book was released on 2009 with total page 124 pages. Available in PDF, EPUB and Kindle. Book excerpt:

Download or read book Advanced Model Predictive Control for Autonomous Marine Vehicles written by Yang Shi and published by Springer Nature. This book was released on 2023-02-13 with total page 210 pages. Available in PDF, EPUB and Kindle. Book excerpt: This book provides a comprehensive overview of marine control system design related to underwater robotics applications. In particular, it presents novel optimization-based model predictive control strategies to solve control problems appearing in autonomous underwater vehicle applications. These novel approaches bring unique features, such as constraint handling, prioritization between multiple design objectives, optimal control performance, and robustness against disturbances and uncertainties, into the control system design. They therefore form a more general framework to design marine control systems and can be widely applied. Advanced Model Predictive Control for Autonomous Marine Vehicles balances theoretical rigor – providing thorough analysis and developing provably-correct design conditions – and application perspectives – addressing practical system constraints and implementation issues. Starting with a fixed-point positioning problem for a single vehicle and progressing to the trajectory-tracking and path-following problem of the vehicle, and then to the coordination control of a large-scale multi-robot team, this book addresses the motion control problems, increasing their level of challenge step-by-step. At each step, related subproblems such as path planning, thrust allocation, collision avoidance, and time constraints for real-time implementation are also discussed with solutions. In each chapter of this book, compact and illustrative examples are provided to demonstrate the design and implementation procedures. As a result, this book is useful for both theoretical study and practical engineering design, and the tools provided in the book are readily applicable for real-world implementation.

Download or read book Optimal Fault Detection and Resolution During Maneuvering for Autonomous Underwater Vehicles written by Andrew S. Gibbons and published by . This book was released on 2000 with total page 148 pages. Available in PDF, EPUB and Kindle. Book excerpt: In order to increase robustness, reliability, and mission success rate, autonomous vehicles must detect debilitating system control faults. Prior model-based observer design for 21UUV was analyzed using actual vehicle sensor data. It was shown, based on experimental response, that residual generation during maneuvering was too excessive to detect manually implemented faults. Optimization of vehicle hydrodynamic coefficients in the model significantly decreased maneuvering residuals, but did not allow for adequate fault detection. Kalman filtering techniques were used to improve residual reduction during maneuvering and increase residual generation during fault conditions. Optimization of the Kalman filter's system noise matrix, measurement noise matrix, and input gain scalar multiplier produced fault resolution which allowed for accurate detection of fault of relatively minor magnitude within minimal time constraints.

Download or read book Exploiting Structure to Efficiently Solve Large Scale Partially Observable Markov Decision Processes microform written by Pascal Poupart and published by Library and Archives Canada = Bibliothèque et Archives Canada. This book was released on 2005 with total page 288 pages. Available in PDF, EPUB and Kindle. Book excerpt: Partially observable Markov decision processes (POMDPs) provide a natural and principled framework to model a wide range of sequential decision making problems under uncertainty. To date, the use of POMDPs in real-world problems has been limited by the poor scalability of existing solution algorithms, which can only solve problems with up to ten thousand states. In fact, the complexity of finding an optimal policy for a finite-horizon discrete POMDP is PSPACE-complete. In practice, two important sources of intractability plague most solution algorithms: Large policy spaces and large state spaces. In practice, it is critical to simultaneously mitigate the impact of complex policy representations and large state spaces. Hence, this thesis describes three approaches that combine techniques capable of dealing with each source of intractability: VDC with BPI, VDC with Perseus (a randomized point-based value iteration algorithm by Spaan and Vlassis [136]), and state abstraction with Perseus. The scalability of those approaches is demonstrated on two problems with more than 33 million states: synthetic network management and a real-world system designed to assist elderly persons with cognitive deficiencies to carry out simple daily tasks such as hand-washing. This represents an important step towards the deployment of POMDP techniques in ever larger, real-world, sequential decision making problems. On the other hand, for many real-world POMDPs it is possible to define effective policies with simple rules of thumb. This suggests that we may be able to find small policies that are near optimal. This thesis first presents a Bounded Policy Iteration (BPI) algorithm to robustly find a good policy represented by a small finite state controller. Real-world POMDPs also tend to exhibit structural properties that can be exploited to mitigate the effect of large state spaces. To that effect, a value-directed compression (VDC) technique is also presented to reduce POMDP models to lower dimensional representations.

Download or read book Robust Model Based Fault Diagnosis for Unmanned Underwater Vehicles Using Sliding Mode Observers written by and published by . This book was released on 1999 with total page 9 pages. Available in PDF, EPUB and Kindle. Book excerpt: The early detection of the malfunctions and faults as well as their compensation is crucial both for maintenance and for mission reliability of unmanned underwater vehicles (UUVs). Among the different fault detection methods using analytical redundancy, the first distinction rises between model-free and model-based approaches. Model-free methods are well-suited for large-scale systems, where the development of a model is too expensive. The lumped parameter model of an underwater vehicle can be easily described by a small set of well-known equations with highly uncertain parameters. This uncertainty suggests the introduction of robustness requirements in the model-based residual generation for UUVs. Robustness can be addressed in many different ways. According to the same view, the so-called unknown input observer (UIO) has been proposed. These approaches share the common idea of decoupling residuals and noises by eliminating (or at least reducing in some optimal sense) the influence of the disturbances on the residuals.

Download or read book Underwater Vehicle Control and Communication Systems Based on Machine Learning Techniques written by Tien Anh Tran and published by . This book was released on 2024 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: "The development of intelligent transportation systems has become significant in marine engineering especially Autonomous Underwater Vehicles with an aim to enhance energy efficiency management and communication systems. This book covers different aspects of optimization autonomous underwater vehicles and their propulsion systems via machine learning techniques. It further analyses hydrodynamic characteristics including study of experimental investigation combined with hydrodynamic characteristics backed my MATLAB codes and simulation study results. Features: Covers utilization of machine learning techniques with a focus on marine science and ocean engineering. Details effect of the intelligent transportation system (ITS) into the sustainable environment and ecology system. Evaluates performance of particle swarm intelligent based optimization techniques. Reviews propulsion performance of the remoted control vehicles based on machine learning techniques. Includes MATLAB examples and simulation study results. This book is aimed at graduate students and researchers in marine engineering and technology, computer science, and control system engineering"--

Download or read book Fault Detection Isolation and Identification of Autonomous Underwater Vehicles Using Dynamic Neural Networks and Genetic Algorithms written by Shaghayegh Shahrokhi Tehrani and published by . This book was released on 2015 with total page 255 pages. Available in PDF, EPUB and Kindle. Book excerpt: The main objective of this thesis is to propose and develop a fault detection, isolation and identification scheme based on dynamic neural networks (DNNs) and genetic algorithm (GA) for thrusters of the autonomous underwater vehicles (AUVs) which provide the force for performing the formation missions. In order to achieve the fault detection task, in this thesis two level of fault detection are proposed, I) Agent-level fault detection (ALFD) and II) Formation-level fault detection (FLFD). The proposed agent-level fault detection scheme includes a dynamic neural network which is trained with absolute measurements and states of each thruster in the AUV. The genetic algorithm is used in order to train the DNN. The results from simulations indicate that although the ALFD scheme can detect the high severity faults, for low severity faults the accuracy is not satisfy our expectations. Therefore, a formation-level fault detection scheme is developed. In the proposed formation-level fault detection scheme, a fault detection unit consist of two dynamic neural networks corresponding to its adjacent neighbors, is employed in each AUV to detect the fault in formation. Each DNN of the fault detection unit is trained with one relative and one absolute measurements. Similar to ALFD scheme, these two DNNs are trained with GA. The simulation results and confusion matrix analysis indicate that our proposed FLFD can detect both low severity and high severity faults with high level of accuracy compare to ALFD scheme. In order to indicate the type and severity of the occurred fault the agent-level and formation-level fault isolation and identification schemes are developed and their performances are compared. In the proposed fault isolation and identification schemes, two neural networks are employed for isolating the type of the fault in the thruster of the AUV and determining the severity of the occurred fault. In the fist step, a multi layer perceptron (MLP) neural network categorize the type of the fault into thruster blocking, flooded thruster and loss of effectiveness in rotor and in the next step a MLP neural network classify the severity into low, medium and high. The neural networks in fault isolation and identification schemes are trained based on genetic algorithm with various data sets which are obtained through different faulty operating condition of the AUV. The simulation results and the confusion matrix analysis indicate that the proposed formation-level fault isolation and identification schemes have a better performance comparing to agent-level schemes and they are capable of isolating and identifying the faults with high level of accuracy and precision.

Download or read book Formation Control and Fault Accommodation for a Team of Autonomous Underwater Vehicles written by Sahar Sedaghati and published by . This book was released on 2015 with total page 188 pages. Available in PDF, EPUB and Kindle. Book excerpt: The purpose of this thesis is the development of efficient formation control and fault accommodation algorithms for a team of autonomous underwater vehicles (AUVs). The team of AUVs are capable of performing a wide range of deep water marine applications such as seabed mapping and surveying, oil and gas exploration and extraction, and oil and gas pipeline inspection. However, communication limitations and the presence of undesirable events such as component faults in any of the team members can prevent the whole team to achieve safe, reliable, and efficient performance while executing underwater mission tasks. In this regard, the semi-decentralized control scheme is developed to achieve trajectory tracking and formation keeping while requiring information exchange only among neighboring agents. To this end, model predictive control (MPC) technique and dynamic game theory are utilized to formulate and solve the formation control problem. Moreover, centralized and decentralized control schemes are developed to assess the performance of the proposed semi-decentralized control scheme in the simulation studies. The simulation results verify that the performance of the proposed semi-decentralized scheme is very close to the centralized scheme with lower control effort cost while it does not impose stringent communication requirements as in the centralized scheme. Moreover, the semi-decentralized active fault recovery scheme is developed to maintain a graceful degraded performance and to ensure that the team of autonomous underwater vehicles can satisfy mission objectives when an actuator fault occurs in any of the team members. In this regard, online fault information provided by fault detection and isolation (FDI) modules of each agent and its neighbors are incorporated to redesign the nominal controllers based on the MPC technique and dynamic game theory. Additionally, FDI imperfections such as fault estimation error and time delay are taken into account, and a performance index is derived to show the impact of FDI imperfections on the performance of team members. Moreover, centralized and decentralized active fault recovery schemes are developed to evaluate the performance of the proposed semi-decentralized recovery scheme through comparative simulation studies with various fault scenarios. The comparative simulation studies justify that the proposed semi-decentralized fault recovery scheme meets the design specifications even if the performance of the FDI module is not ideal.

Download or read book Cooperative Control and Fault Recovery for Network of Heterogeneous Autonomous Underwater Vehicles written by Maria Enayat and published by . This book was released on 2017 with total page 249 pages. Available in PDF, EPUB and Kindle. Book excerpt: The purpose of this thesis is to develop cooperative recovery control schemes for a team of heterogeneous autonomous underwater vehicles (AUV). The objective is to have the network of autonomous underwater vehicles follow a desired trajectory while agents maintain a desired formation. It is assumed that the model parameters associated with each vehicle is different although the order of the vehicles are the same. Three cooperative control schemes based on dynamic surface control (DSC) technique are developed. First, a DSC-based centralized scheme is presented in which there is a central controller that has access to information of all agents at the same time and designs the optimal solution for this cooperative problem. This scheme is used as a benchmark to evaluate the performance of other schemes developed in this thesis. Second, a DSC-based decentralized scheme is presented in which each agent designs its controller based on only its information and the information of its desired trajectory. In this scheme, there is no information exchange among the agents in the team. This scheme is also developed for the purpose of comparative studies. Third, two different semi-decentralized or distributed schemes for the network of heterogeneous autonomous underwater vehicles are proposed. These schemes are a synthesis of a consensus-based algorithm and the dynamic surface control technique with the difference that in one of them the desired trajectories of agents are used in the consensus algorithm while in the other the actual states of the agents are used. In the former scheme, the agents communicate their desired relative distances with the agents within their set of nearest neighbors and each agent determines its own control trajectory. In this semi-decentralized scheme, the velocity measurements of the virtual leader and all the followers are not required to reach the consensus formation. However, in the latter, agents communicate their relative distances and velocities with the agents within their set of nearest neighbors. In both semi-decentralized schemes only a subset of agents has access to information of a virtual leader. The comparative studies between these two semi-decentralized schemes are provided which show the superiority of the former semi-decentralized scheme over latter. Furthermore, to evaluate the efficiency of the proposed DSC-based semi-decentralized scheme with consensus algorithm using desired trajectories, a comparative study is performed between this scheme and three cooperative schemes of model-dependent coordinated tracking algorithm, namely the centralized, decentralized, and semi-decentralized schemes. Given that the dynamics of autonomous underwater vehicles are inevitably subjected to system faults, and in particular the actuator faults, to improve the performance of the network of agents, active fault-tolerant control strategies corresponding to the three developed schemes are also designed to recover the team from the loss-of-effectiveness in the actuators and to ensure that the closed-loop signals remain bounded and the team of heterogeneous autonomous underwater vehicles satisfy the overall design specifications and requirements. The results of this research can potentially be used in various marine applications such as underwater oil and gas pipeline inspection and repairing, monitoring oil and gas pipelines, detecting and preventing any oil and gas leakages. However, the applications of the proposed cooperative control and its fault-tolerant scheme are not limited to underwater formation path-tracking and can be applied to any other multi-vehicle systems that are characterized by Euler-Lagrange equations.

Download or read book Fault Tolerant Control of Autonomous Underwater Vehicles written by Douglas Edward Perrault and published by . This book was released on 1998 with total page 246 pages. Available in PDF, EPUB and Kindle. Book excerpt:

Download or read book Algorithms for Partially Observable Markov Decision Processes written by Weihong Zhang and published by . This book was released on 2001 with total page 374 pages. Available in PDF, EPUB and Kindle. Book excerpt:

Download or read book Fault tolerant Synchronization of Autonomous Underwater Vehicles written by Faegheh Amirarfaei and published by . This book was released on 2016 with total page 210 pages. Available in PDF, EPUB and Kindle. Book excerpt: The main objective of this thesis is to develop a fault-tolerant and reconfigurable synchronization scheme based on model-based control protocols for stern and sail hydroplanes that are employed as actuators in the attitude control subsystem (ACS) of an autonomous underwater vehicle (AUV). In this thesis two control approaches are considered for synchronization, namely i) state feedback synchronization, and ii) output feedback synchronization. Both problems are tackled by proposing a passive control approach as well as an active reconfiguration (re-designing the control gains). For the ?state feedback? synchronization scheme, to achieve consensus the relative/absolute measurements of the AUV’s states (position and attitude) are available. The states of a longitudinal model of an AUV are mainly heave, pitch, and their associated rates. For the state feedback problem we employ a static protocol, and it is shown that the multi-agent system will synchronize in the stochastic mean square sense in the presence of measurement noise. However, the resulting performance index defined as the accumulated sum of variations of control inputs and synchronization errors is high. To deal with this problem, Kalman filtering is used for states estimation that are used in synchronization protocol. Moreover, the e�ffects of parameter uncertainty of the agent’s dynamics are also investigated through simulation results. By employing the static protocol it is demonstrated that when a loss of e�ffectiveness (LOE) or float fault occurs the synchronization can still be achieved under some conditions. Finally, one of the main problems that is tackled in the state feedback scenario is our proposed proportional-integral (PI) control methodology to deal with the lock in place (LIP) fault. It is shown that if the LIP fault occurs, by employing a PI protocol the synchronization could still be achieved. Finally, our proposed dynamic synchronization protocol methodology is applied given that the fault (LOE/float) severity is known. Since after a fault occurrence the agents become heterogeneous, employing the dynamic scheme makes the task of reconfiguration (redesigning the gains) more e�ffective. For the ?output feedback? synchronization approach, to achieve consensus relative/absolute measurements of the AUV’s states except the pitch rate are available. For the output feedback problem a dynamic protocol through a Luenberger observer is first employed for state estimation and the synchronization achievement is demonstrated. Then, a system under state and measurement noise is considered, and it is shown that by employing a Kalman filter for the state estimation; the multi-agent system will synchronize in the stochastic mean square sense. Furthermore, by employing the static protocol, it is shown that when a LOE/float fault occurs the synchronization is still achieved under certain conditions. Finally, one of the main problems that is tackled in the output feedback scenario is our proposed dynamic controller methodology. The results of this scheme are compared with another approach that exploits both dynamic controller and dynamic observer. The former approach has less computational e�ort and results in more a robust control with respect to the actuator fault. The reason is that the later method employs an observer that uses the control input matrix information. When fault occurs, this information will not be correct any more. However, if there is a need to redesign the synchronization gains under faulty scenario, the later methodology is preferred. The reason is that the former approach becomes complicated when there is a fault even though its severity is known. In this thesis, fault-tolerant synchronization of autonomous underwater vehicles is considered. In the first chapter a brief introduction on the motivation, problem definition, objectives and the methodologies that are used in the dissertation are discussed. A literature review on research dedicated to synchronization, fault diagnosis, and fault-tolerant control is provided. In Chapter 2, a through literature review on unmanned underwater vehicles is covered. It also comprises a comprehensive background information and definitions including algebraic graph theory, matrix theory, and fault modeling. In the problem statement, the two main problems in this thesis, namely state feedback synchronization and output feedback synchronization are discussed. Chapters 3 and 4 will cover these two problems, their solutions, and the corresponding simulation results that are provided. Finally, Chapter 5 includes a discussion of conclusions and future work.

Download or read book Exact and Approximate Algorithms for Partially Observable Markov Decision Processes written by Anthony Rocco Cassandra and published by . This book was released on 1998 with total page 894 pages. Available in PDF, EPUB and Kindle. Book excerpt:

Download or read book Active Learning in Partially Observable Markov Decision Processes written by Robin Jaulmes and published by . This book was released on 2006 with total page 200 pages. Available in PDF, EPUB and Kindle. Book excerpt: "After reviewing existing methods for solving learning problems in partially observable environments, we expose a theoretical active learning setup. We propose an algorithm, MEDUSA, and show theoretical and empirical proofs of performance for it." --

Download or read book Dynamic Programming for Partially Observable Markov Decision Processes Using a Factored Representation written by Zhengzhu Feng and published by . This book was released on 2000 with total page 116 pages. Available in PDF, EPUB and Kindle. Book excerpt:

Download or read book Decision Making Under Uncertainty written by Mykel J. Kochenderfer and published by MIT Press. This book was released on 2015-07-24 with total page 350 pages. Available in PDF, EPUB and Kindle. Book excerpt: An introduction to decision making under uncertainty from a computational perspective, covering both theory and applications ranging from speech recognition to airborne collision avoidance. Many important problems involve decision making under uncertainty—that is, choosing actions based on often imperfect observations, with unknown outcomes. Designers of automated decision support systems must take into account the various sources of uncertainty while balancing the multiple objectives of the system. This book provides an introduction to the challenges of decision making under uncertainty from a computational perspective. It presents both the theory behind decision making models and algorithms and a collection of example applications that range from speech recognition to aircraft collision avoidance. Focusing on two methods for designing decision agents, planning and reinforcement learning, the book covers probabilistic models, introducing Bayesian networks as a graphical model that captures probabilistic relationships between variables; utility theory as a framework for understanding optimal decision making under uncertainty; Markov decision processes as a method for modeling sequential problems; model uncertainty; state uncertainty; and cooperative decision making involving multiple interacting agents. A series of applications shows how the theoretical concepts can be applied to systems for attribute-based person search, speech applications, collision avoidance, and unmanned aircraft persistent surveillance. Decision Making Under Uncertainty unifies research from different communities using consistent notation, and is accessible to students and researchers across engineering disciplines who have some prior exposure to probability theory and calculus. It can be used as a text for advanced undergraduate and graduate students in fields including computer science, aerospace and electrical engineering, and management science. It will also be a valuable professional reference for researchers in a variety of disciplines.