检索结果-内蒙古大学图书馆

6th Asia Conference on Machine Learning and Computing, ACMLC 2024

作者： Monko, Gloriana J. Kimura, Masaomi Department of Functional Control Systems Shibaura Institute of Technology Tokyo Japan School of Engineering Shibaura Institute of Technology Koto City Tokyo Japan

ISBN: (纸本)9798400710018

This study introduces a novel method for integrating Stratified Sampling for Density-Based Spatial Clustering of Applications with Noise (SS-DBSCAN) clustering with the human-in-the-loop approach to semi-supervised data labeling. We enhance the feature representation of text data using attention mechanisms, leading to the formation of more contextually relevant clusters. The clusters are then processed using SS-DBSCAN, where a custom similarity function aligns them closely with expert-labeled examples, facilitating efficient label propagation. Subsequently, human experts reevaluate the clustering outputs, refining the process by reassessing the feature similarities to the expert-labeled classes. This iterative process ensures high fidelity in label assignment and enhances the overall reliability of the semi-supervised learning approach. Applied to MIMIC III and Cancer Doc Classification datasets, this method not only aligns the data labels more closely with expert evaluations but also leverages human judgment to confirm the accuracy and reliability of the re-clustering process. Based on the BERT transformers model, the fine-tuned prediction model achieved 100% accuracy in predicting the correct label and evaluated the re-clustering technique's effectiveness, demonstrating 90.3% and 97.8% correctness in the re-clustered labels in MIMIC III and Cancer Doc Classification datasets, respectively. © 2024 Copyright held by the owner/author(s).

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Modified Covariance Intersection-Based Sensor Fault Tolerance of a Gas Turbine Engine

Modified Covariance Intersection-Based Sensor Fault Toleranc...

引用

AIAA Science and technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Chakraborty, Tirthankar Maity, Arnab Intelligent Systems and Control Lab Department of Aerospace Engineering Indian Institute of Technology Bombay Mumbai400076 India

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

Fault tolerance is crucial in the design and operation of gas turbine engines, where even minor malfunctions can disrupt operations and compromise performance. This paper presents a modified covariance intersection-based decentralized estimator which can ensure uninterrupted operation amidst sensor failures. A network of sensor nodes within the engine is considered. Each node in the network computes a local state estimate, which is subsequently fused with the other node estimates at each time instant. The developed algorithm uses a novel weight adaptation scheme to fuse the node estimates and the associated covariances. The node weights are expressed as a function of the covariance matrix of the node estimate and a scaling matrix. This algorithm is tested through simulation by injecting faults in the sensors of a small-scale gas turbine engine primarily used for research, called the SR-30 engine. The simulation results illustrate the effectiveness of the proposed estimator in ensuring the integrity of the state estimates in the presence of drift and bias faults in the sensors. © 2025, American institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Gas turbines

来源：评论

学校读者我要写书评

暂无评论

Nested Sliding Mode control Based Impact Angle Guidance with Bounded Input

Nested Sliding Mode Control Based Impact Angle Guidance with...

引用

AIAA Science and technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Surve, Prajakta Kumar, Shashi Ranjan Intelligent Systems & Control Lab Department of Aerospace Engineering Indian Institute of Technology Bombay Powai Mumbai400076 India

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

In this paper, a nonlinear guidance law for impact angle-constrained interception of nonmaneuvering targets, respecting the bounds on the control input - the commanded interceptor lateral acceleration, is proposed. By modeling the required interceptor lateral acceleration as a first-order dynamics, which is a function of the maximum permissible value of control input and guidance command, the control input is maintained within the predefined bounds. The dynamics is chosen in such a way that the resulting input will never violate the prescribed bounds. To design the guidance command, also known as commanded lateral acceleration, we employ a sliding surface based on the error between actual and desired line-of-sight (LOS) angles and the LOS rate, facilitating a collision course with zero miss distance. A secondary sliding surface is designed as a function of the primary sliding surface and its rate. This sliding surface enables the control over the required interceptor lateral acceleration. An effective guidance command for precise interception at given impact angle, considering the bounds on control input, is then derived using the sliding mode control technique. Numerical simulations validate the merits and claims of the presented guidance approach. © 2025, American institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

CONVERGENCE GUARANTEES OF MODEL-FREE POLICY GRADIENT METHODS FOR LQR WITH STOCHASTIC DATA

arXiv

引用

arXiv 2025年

作者： Song, Bowen Iannelli, Andrea Institute for Systems Theory and Automatic Control University of Stuttgart Germany

Policy gradient (PG) methods are the backbone of many reinforcement learning algorithms due to their good performance in policy optimization problems. As a gradient-based approach, PG methods typically rely on knowledge of the system dynamics. However, this information is not always available, and in such cases, trajectory data can be utilized to approximate first-order information. When the data are noisy, gradient estimates become inaccurate and a formal investigation that encompasses uncertainty estimation and the analysis of its propagation through the algorithm is currently missing. To address this, our work focuses on the Linear Quadratic Regulator (LQR) problem for systems subject to additive stochastic noise. After briefly summarizing the state of the art for cases with a known model, we focus on scenarios where the system dynamics are unknown, and approximate gradient information is obtained using zeroth-order optimization techniques. We analyze the theoretical properties by computing the error in the estimated gradient and examining how this error affects the convergence of PG algorithms. Additionally, we provide global convergence guarantees for various versions of PG methods, including those employing adaptive step sizes and variance reduction techniques, which help increase the convergence rate and reduce sample complexity. One contribution of this work is the study of the robustness of model-free PG methods, aiming to identify their limitations in the presence of noise and propose improvements to enhance their applicability. Numerical simulations show that these theoretical analyses provide valuable guidance in tuning the algorithm parameters, thereby making these methods more reliable in practically relevant scenarios. © 2025, CC BY.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Adaptive Unscented Kalman Filter Based Sensor Fusion for Aircraft Positioning Relative to Unknown Runway

Adaptive Unscented Kalman Filter Based Sensor Fusion for Air...

引用

AIAA Science and technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Ghosh, Nabarun Maity, Arnab Intelligent Systems and Control Lab Department of Aerospace Engineering Indian Institute of Technology Bombay Maharashtra Mumbai400076 India

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

This paper proposes an estimator to provide precise position, velocity, and orientation for a landing aircraft relative to the runway based on image sensors, IMU sensor data, and barometric sensor measurements. This work excludes GPS data, as it is expected that the proposed estimator can be used for fault detection in GPS and instrument landing systems. The runway is unknown, and hence, its width is estimated online. An adaptive UKF-based approach is adopted to estimate the states. In order to verify the nearer-to-practical performance and accuracy of bias estimation, realistically simulated data that constitute noises and biases from the sensors are taken into consideration. The proposed approach provides a solution for nonlinear state estimation in the presence of measurement noises that are non-additive. In addition to this, it provides estimates of the unknown measurement noise covariance. A comparative study is conducted between the conventional UKF and the proposed AUKF to analyze the adopted approach. Results show that the proposed algorithm gives comparable estimation results and faster convergence than the conventional UKF. © 2025, American institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Fault detection

来源：评论

学校读者我要写书评

暂无评论

Measurement of SPT-140 Hall Thruster Swirl Torque during the Psyche Mission

Measurement of SPT-140 Hall Thruster Swirl Torque during the...

引用

AIAA Science and technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Snyder, John Steven Nicholas, Austin K. Collins, Steven M. Ortega, Alejandro Lopez Electric Propulsion Group Jet Propulsion Laboratory California Institute of Technology PasadenaCA United States Navigation and Control Systems Engineer Jet Propulsion Laboratory California Institute of Technology PasadenaCA United States Guidance and Control Systems Section Jet Propulsion Laboratory California Institute of Technology PasadenaCA United States

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

Hall-thruster swirl torque is small enough that it is difficult to measure in ground-based facilities, but large enough that its effects on interplanetary spacecraft can be significant and must be actively managed. NASA’s Psyche mission has provided the first opportunity to directly measure the swirl torque produced by a Hall thruster operating at powers greater than 1.5 kW. When the electric propulsion system is operating, the spacecraft attitude control methodology forces the net torque on the spacecraft to align with a pre-determined direction in the body frame, after which the thruster-induced swirl torque can be separated from the other torques acting on the spacecraft. The swirl torques produced by three of the four thrusters on the Psyche spacecraft were determined over the full operating discharge power range of 1.0 to 4.5 kW, and the maximum value of that torque was 0.86 mN-m at full power. It is shown that the SPT-140 swirl torque results can be scaled from full power to lower powers by the product of the discharge current and the magnet current, and evaluation of cruise thrusting data as the thruster is throttled down in power suggests that swirl torque is not changing significantly with time as a function of operating condition. Finally, calculations of swirl torque using updated numerical simulations are shown to agree reasonably well with the flight data, thus providing confidence in their use for other thrusters prior to flight. © 2025, American institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： NASA

来源：评论

学校读者我要写书评

暂无评论

Robust Adaptive Sliding Mode Security control of Markov Jump Cyber-Physical systems With Stochastic Injection Attacks Through Event-Triggered-Based Observer Approach

引用

IEEE Transactions on systems, Man, and Cybernetics: systems 2025年第5期55卷 3679-3692页

作者： Jiang, Baoping Niu, Fuzhou Wu, Zhengtian Qiu, Jianbin Suzhou University of Science and Technology School of Electronic and Information Engineering Suzhou215009 China Politecnico di Milano Department of Mechanical Engineering Milan20156 Italy Research Institute of Intelligent Control and Systems Harbin Institute of Technology Harbin150001 China

This article addresses the challenge of state observer design for sliding mode security control in Markov jump cyber-physical systems subjected to stochastic injection attacks. To enhance network efficiency, a dynamic event-triggered algorithm is introduced in the communication channel. First, the design begins with a Luenberger state observer featuring an adaptive compensator. This configuration aims to effectively counteract malicious attacks. Second, an integral sliding hyperplane is formulated within the estimation space, which serves as the foundation for deriving the sliding mode dynamics, ensuring robustness against disturbances. Recognizing the diversity of transition rates (TRs), an elastic sliding mode controller is designed to accommodate three distinct types of TRs, which is also strategically designed to guarantee reachability and maintain sliding motion. Third, stochastic stability with an H∞ attenuation level is conducted separately for each type of TR. Correspondingly, the development of an algorithm for determining threshold parameters in triggered conditions is presented. Simultaneously, a proof of the nonexistence of Zeno behavior is provided, ensuring the stability and efficiency of the proposed system. Finally, a simulation study using a practical model is included to empirically demonstrate the validity of the proposed method in a real-world context. © 2013 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

An interaction-fair semi-decentralized trajectory planner for connected and autonomous vehicles

引用

Autonomous Intelligent systems 2025年第1期5卷 1-20页

作者： Liu, Zhengqin Lei, Jinlong Yi, Peng Hong, Yiguang Department of Control Science and Engineering Tongji University Shanghai201804 China The Shanghai Research Institute for Intelligent Autonomous Systems Shanghai201804 China Shanghai Institute of Intelligent Science and Technology Tongji University Shanghai201804 China

Lately, there has been a lot of interest in game-theoretic approaches to the trajectory planning of autonomous vehicles (AVs). But most methods solve the game independently for each AV while lacking coordination mechanisms, and hence result in redundant computation and fail to converge to the same equilibrium, which presents challenges in computational efficiency and safety. Moreover, most studies rely on the strong assumption of knowing the intentions of all other AVs. This paper designs a novel autonomous vehicle trajectory planning approach to resolve the computational efficiency and safety problems in uncoordinated trajectory planning by exploiting vehicle-to-everything (V2X) technology. Firstly, the trajectory planning for connected and autonomous vehicles (CAVs) is formulated as a game with coupled safety constraints. We then define the interaction fairness of the planned trajectories and prove that interaction-fair trajectories correspond to the variational equilibrium (VE) of this game. Subsequently, we propose a semi-decentralized planner for the vehicles to seek VE-based fair trajectories, in which each CAV optimizes its individual trajectory based on neighboring CAVs’ information shared through V2X, and the roadside unit takes the role of updating multipliers for collision avoidance constraints. The approach can significantly improve computational efficiency through parallel computing among CAVs, and enhance the safety of planned trajectories by ensuring equilibrium concordance among CAVs. Finally, we conduct Monte Carlo experiments in multiple situations at an intersection, where the empirical results show the advantages of SVEP, including the fast computation speed, a small communication payload, high scalability, equilibrium concordance, and safety, making it a promising solution for trajectory planning in connected traffic scenarios. To the best of our knowledge, this is the first study to achieve semi-distributed solving of a game with coupled constr

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Fault-Tolerant Soft Sensor Modeling Based on a Two-Dimensional Group Distributionally Robust Optimization Framework

引用

IEEE Transactions on Automation Science and Engineering 2025年 22卷 14396-14406页

作者： Zhang, Xiangrui Song, Chunyue Zhao, Jun Huang, Biao College of Control Science and Engineering Zhejiang University State Key Laboratory of Industrial Control Technology Institute of Industry Intelligence and Systems Engineering Hangzhou310027 China University of Alberta Department of Chemical and Materials Engineering EdmontonABT6G 1H9 Canada

In industrial automation and intelligence, fault tolerance mechanisms have always been an attractive topic. To develop soft sensors with fault tolerance for different types of faults and unforeseen new faults, this article proposes a two-dimensional group distributionally robust optimization (2D-GDRO) framework for fault-tolerant soft sensor modeling. We propose to describe the potential distributions of new fault conditions with an uncertainty set and optimize the soft sensor model by minimizing the worst-case risk over the uncertainty set. Considering the restricted representation range of the uncertainty set constructed directly from a mixture distribution of a limited number of existing fault conditions in the training set, a two-dimensional uncertainty set is designed at the group dimension and the sample dimension. To efficiently train a fault-tolerant soft sensor within the 2D-GDRO framework, we introduce a triple-interleaved optimization algorithm. This algorithm integrates mini-batch stochastic gradient descent, exponentiated gradient ascent, and group-wise SoftMax techniques. Finally, the fault tolerance of the 2D-GDRO framework based soft sensor is verified using the Tennessee-Eastman process and the real three-phase flow facility. The experimental results show that 2D-GDRO outperforms other training frameworks in average soft sensing accuracy under new fault conditions. © 2004-2012 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Multi-objective robust controller synthesis with integral quadratic constraints in discrete-time

arXiv

引用

arXiv 2025年

作者： Schwenkel, Lukas Köhler, Johannes Müller, Matthias A. Scherer, Carsten W. Allgöwer, Frank University of Stuttgart Institute for Systems Theory and Automatic Control Germany ETH Zürich Institute for Dynamical Systems and Control Switzerland Leibniz University Hannover Institute for Automatic Control Germany University of Stuttgart Department of Mathematics Germany

This article presents a novel framework for the robust controller synthesis problem in discrete-time systems using dynamic Integral Quadratic Constraints (IQCs). We present an algorithm to minimize closed-loop performance measures such as the H∞-norm, the energy-to-peak gain, the peak-to-peak gain, or a multi-objective mix thereof. While IQCs provide a powerful tool for modeling structured uncertainties and nonlinearities, existing synthesis methods are limited to the H∞-norm, continuous-time systems, or special system structures. By minimizing the energy-to-peak and peak-to-peak gain, the proposed synthesis can be utilized to bound the peak of the output, which is crucial in many applications requiring robust constraint satisfaction, input-to-state stability, reachability analysis, or other pointwise-in-time bounds. Numerical examples demonstrate the robustness and performance of the controllers synthesized with the proposed algorithm. Copyright © 2025, The Authors. All rights reserved.

关键词： Robust control

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：