检索结果-内蒙古大学图书馆

2024 International Conference on Computing, Robotics and System Sciences, ICRSS 2024

作者： Wei, Cuihua Guangdong University of Science and Technology Guangdong Province Dongguan City523668 China

ISBN: (纸本)9798331511296

With the widespread application of robot technology in target recognition and intelligent localization, issues such as slow convergence and low accuracy in traditional BP neural networks and Particle Swarm optimization (PSO) algorithms have gradually become prominent. To address these problems, an improved PSO-BP algorithm is proposed, which optimizes PSO by incorporating chaotic sequences and adaptive strategies. Simultaneously, PSO is utilized to optimize the initial weights and biases of the BP neural network, enhancing recognition accuracy and localization precision. Experimental results demonstrate that the improved PSO-BP algorithm exhibits higher robustness and stability under various lighting and noise conditions. In an environment with 300 lux lighting and low noise, the recognition accuracy of this algorithm reaches 92.5%, with a localization error of 1.12 cm, significantly outperforming traditional algorithms. Furthermore, the algorithm reduces the error to 0.05 within 100 iterations, showcasing rapid convergence. In summary, the improved PSO-BP algorithm demonstrates high accuracy, stability, and efficiency in robot target recognition and localization, providing an effective solution for robot applications in complex environments. © 2024 IEEE.

关键词： Adaptive strategy Intelligent positioning Intelligent system optimization algorithm Robotics Target recognition

来源：评论

学校读者我要写书评

暂无评论

Federated Learning with Class Balanced Loss Optimized by Implicit Stochastic Gradient Descent 7th

Federated Learning with Class Balanced Loss Optimized by Imp...

引用

7th International Conference on Soft Computing in Data Science (SCDS)

作者： Zhou, Jincheng Zheng, Maoxing Qiannan Normal Univ Nationalities Sch Comp & Informat Duyun 558000 Peoples R China Baoji Univ Arts & Sci Sch Comp Sci Baoji 721007 Peoples R China Key Lab Complex Syst & Intelligent Optimizat Guiz Duyun 558000 Peoples R China

ISBN: (纸本)9789819904044;9789819904051

Federated learning is a paradigm for distributed machine learning in which a central server interacts with a large number of remote devices to create the optimal global model. System and data heterogeneity are now the two largest impediments to federated learning. This work suggests a federated learning strategy based on stochastic gradient descent optimization as a solution to the problem of heterogeneity-induced slow convergence, or even non-convergence, of the global model. This work estimates the average global gradient using locally uploaded model parameters without computing the first derivative or updating global model parameters through gradient descent. Allowing the global model to be used with fewer communication rounds. Obtain faster and more reliable convergence results. In experiments simulating varying degrees of heterogeneous settings, the strategy proposed in this work delivered faster and more stable convergence than FedProx and FedAvg. This work offers a strategy that decreases the number of communication cycles on highly heterogeneous synthetic datasets by around 50% compared to FedProx, therefore considerably enhancing the stability and durability of federated learning.

关键词： Fast convergence optimization algorithm System heterogeneity Data heterogeneity Implicit stochastic gradient descent Global model Central server Distributed machine learning Federated learning

来源：评论

学校读者我要写书评

暂无评论

An Indirect Impedance Eduction Process for Liners with Arbitrarily Complex Geometry

An Indirect Impedance Eduction Process for Liners with Arbit...

引用

AIAA Aviation Forum

作者： Cerizza, Davide Casalinot, Damiano Dassault Syst SIMULIA SIMULIA R&D Fluids Sci & Technol Waltham MA 02451 USA Dassault Syst Deutschland GmbH SIMULIA R&D Fluids Sci & Technol D-70563 Stuttgart Germany

ISBN: (数字)9781624107047

ISBN: (纸本)9781624107047

Complex aero-engine liner geometries that extend the capabilities of standard 1- or 2degree-of-freedom liners became very popular in recent years. The challenge when trying to characterize these devices in terms of an equivalent impedance is that the classical in-situ method to educe impedance from measured pressure fluctuations cannot be employed straightforwardly. The current study proposes an impedance eduction method based on a combined usage of a 3D scale-resolving flow simulation of a limited portion of the liner with a 2D frequency-domain FEM acoustic calculation in which the liner is replaced by an equivalent impedance boundary condition. Both simulations setups are generated by a fully automatic workflow based on specific user settings. The FEM calculations are driven by an optimization algorithm that changes the value of wall impedance until acoustic properties of the reference 3D solution are matched.

关键词： Finite Element Method Acoustic Propagation Frequency Domain Aircraft Engines optimization algorithm Flow Conditions Boundary Layer Thickness Velocity Profiles Large Eddy Simulation Grazing Flow Impedance Tube

来源：评论

学校读者我要写书评

暂无评论

Image-Based Meta-Reinforcement Learning for Autonomous Guidance of an Asteroid Impactor

引用

JOURNAL OF GUIDANCE CONTROL AND DYNAMICS 2022年第11期45卷 2013-2028页

作者： Federici, Lorenzo Scorsoglio, Andrea Ghilardi, Luca D'Ambrosio, Andrea Benedikter, Boris Zavoli, Alessandro Furfaro, Roberto Sapienza Univ Rome Dept Mech & Aerosp Engn Via Eudossiana 18 I-00184 Rome Italy Univ Arizona Dept Syst & Ind Engn 1127E James E Rogers Way Tucson AZ 85721 USA Sapienza Univ Rome Sch Aerosp Engn Via Salaria 851 I-00138 Rome Italy Univ Arizona Dept Aerosp & Mech Engn 1130 N Mt Ave Tucson AZ 85721 USA

This paper focuses on the use of meta-reinforcement learning for the autonomous guidance of a spacecraft during the terminal phase of an impact mission toward a binary asteroid system. The control policy is replaced by a convolutional-recurrent neural network, which is used to map optical observations collected by the onboard camera to the control thrust and thrusting times. The network is trained by a proximal policy optimization algorithm, a family of reinforcement learning methods. The final phase of NASA's Double Asteroid Redirection Test (DART) mission is used as a test case. The objective is to maneuver the spacecraft to impact the smaller object, Dimorphos, in the Didymos binary system. The spacecraft dynamics are described using the bi-elliptic restricted four-body problem with solar radiation pressure. The initial conditions are randomly scattered according to the actual specifications of the DART mission. A random error on the orbital position of Dimorphos is also considered to reflect uncertainty on the binary system's characteristics and dynamics. The control system aims at minimizing the error on the final spacecraft position. Numerical results show that the guidance system can correctly drive the spacecraft toward the final impact point in more than 98% of the 500 test scenarios.

关键词： Reinforcement Learning Potentially Hazardous Object Deep Convolutional Neural Network Markov Decision Process Guidance System optimization algorithm Spacecraft Dynamics Solar Radiation Interplanetary Trajectories Terminal Guidance

来源：评论

学校读者我要写书评

暂无评论

Aircraft Mathematical Model Identification for Flight Trajectories and Performance Analysis in Cruise

引用

JOURNAL OF AEROSPACE INFORMATION SYSTEMS 2022年第8期19卷 530-549页

作者： Ghazi, Georges Botez, Ruxandra Mihaela Ecole Technol Super ETS Lab Appl Res Act Control Avion & Aeroservoelas Montreal PQ H3C 1K3 Canada

In this paper, a mathematical model for estimating the performance and flight trajectories in cruise is identified from data available in flight manuals. The first part of this paper focuses on the design of a fuel flow and emissions model. Starting from the equations of motion of an aircraft in cruise, a simplified model representing the fuel flow in a corrected form was developed. A practical algorithm was next developed to identify the aircraft model parameters and to determine the mathematical structure that reflects its fuel flow. This process was done using performance data available in the aircraft flight crew operating manual. The emissions model was also developed based on data available in the International Civil Aviation Organization's engine emissions databank. The second part of the paper deals with the development of algorithms for predicting the trajectories and calculating the optimal speeds (i.e., maximum range, long range, and economy) of the aircraft in cruise. Practical techniques for storing and retrieving information without using optimization algorithms have been considered. The methodology was applied on both a Cessna Citation X business jet and Bombardier CRJ-700 regional jet aircraft. The comparison results showed a very good agreement for the fuel consumption and optimal speed.

关键词： Mathematical Models Flight Trajectory Commercial Aircraft Flight Crew Operating Manual Fuel Consumption Cessna Citation X European Organisation for the Safety of Air Navigation Business Jets Aircrafts optimization algorithm Aerodynamic Characteristics

来源：评论

学校读者我要写书评

暂无评论

Diagnosis of Alzheimer's Disease Based on Accelerated Mirror Descent optimization and a Three-Dimensional Aggregated Residual Network

引用

SENSORS 2023年第21期23卷 8708-8708页

作者： Tu, Yue Lin, Shukuan Qiao, Jianzhong Zhang, Peng Hao, Kuankuan Northeastern Univ Dept Comp Sci & Engn Shenyang 110819 Peoples R China

Alzheimer's disease (AD), a neuropsychiatric disorder, continually arises in the elderly. To date, no targeted medications have been developed for AD. Early and fast diagnosis of AD plays a pivotal role in identifying potential AD patients, enabling timely medical interventions, and mitigating disease progression. Computer-aided diagnosis (CAD) becomes possible with the burgeoning of deep learning. However, the existing CAD models for processing 3D Alzheimer's disease images usually have the problems of slow convergence, disappearance of gradient, and falling into local optimum. This makes the training of 3D diagnosis models need a lot of time, and the accuracy is often poor. In this paper, a novel 3D aggregated residual network with accelerated mirror descent optimization is proposed for diagnosing AD. First, a novel unbiased subgradient accelerated mirror descent (SAMD) optimization algorithm is proposed to speed up diagnosis network training. By optimizing the nonlinear projection process, our proposed algorithm can avoid the occurrence of the local optimum in the non-Euclidean distance metric. The most notable aspect is that, to the best of our knowledge, this is the pioneering attempt to optimize the AD diagnosis training process by improving the optimization algorithm. Then, we provide a rigorous proof of the SAMD's convergence, and the convergence of SAMD is better than any existing gradient descent algorithms. Finally, we use our proposed SAMD algorithm to train our proposed 3D aggregated residual network architecture (ARCNN). We employed the ADNI dataset to train ARCNN diagnostic models separately for the AD vs. NC task and the sMCI vs. pMCI task, followed by testing to evaluate the disease diagnostic outcomes. The results reveal that the accuracy can be improved in diagnosing AD, and the training speed can be accelerated. Our proposed method achieves 95.4% accuracy in AD diagnosis and 79.9% accuracy in MCI diagnosis;the best results contrasted with several

关键词： Alzheimer's disease optimization algorithm mirror descent CNN MRI

来源：评论

学校读者我要写书评

暂无评论

Novel Saturated Flow Boiling Heat Transfer Correlation for R32 Refrigerant

引用

JOURNAL OF THERMOPHYSICS AND HEAT TRANSFER 2022年第1期36卷 215-220页

作者： Turgut, Oguz Emrah Asker, Mustafa Genceli, Hadi Coban, Mustafa Turhan Izmir Balarcay Univ Dept Ind Engn Fac Engn & Architecture Gazi Mustafa Kemal St TR-35665 Izmir Turkey Adnan Menderes Univ Dept Mech Engn TR-09010 Aydin Turkey Yildiz Tech Univ Fac Mech Engn TR-34349 Istanbul Turkey Ege Univ Mech Engn Dept Fac Engn TR-35040 Izmir Turkey

来源：评论

学校读者我要写书评

暂无评论

ROT8: A Matlab App for Low Reynolds number Airfoil and Rotor Aerodynamic Design

ROT8: A Matlab App for Low Reynolds number Airfoil and Rotor...

引用

AIAA Aviation Forum

作者： Ruiz, M. Carreno D'Ambrosio, D. Politecn Torino Dept Mech & Aerosp Engn Cso Duca Abruzzi 24 I-10124 Turin Italy

ISBN: (数字)9781624107047

ISBN: (纸本)9781624107047

The proposed software aims to solve the increase in demand for rotor design modules generated by the exponential growth of applications involving Unmanned Aerial Systems (UASs). We offer a variety of aerodynamic solvers for rotor flows with different fidelities, including a Blade Element Momentum (BEM) method, a Free Vortex Wake (FVW) method, a Vortex Particle Method (VPM), and a data-driven approach based on experimental data ROT8 allows the calculation of airfoil polars necessary for the aforementioned Reduced Order Models (ROMs) using XFOIL. This general framework can be applied to rotors, propellers, and horizontal-axis wind turbines. The code is a Matlab App that allows using different optimization algorithms for design purposes. The code is intended to be versatile, with straightforward default settings for academic use but allowing the definition of user-defined functions to increase the level of interaction with the code for advanced research purposes.

关键词： Airfoil Databases MATLAB Aerodynamic optimization Horseshoe Vortex Reduced Order Modelling Propeller Performance optimization algorithm Horizontal Axis Wind Turbine Unmanned Aerial Systems Aerodynamic Efficiency

来源：评论

学校读者我要写书评

暂无评论

A multi-objective optimization of electric vehicles energy flows: the charging process

引用

ANNALS OF OPERATIONS RESEARCH 2021年第1-2期296卷 315-333页

作者： Merhy, Ghimar Nait-Sidi-Moh, Ahmed Moubayed, Nazih Univ Picardie Jules Verne LTI Lab St Quentin en Yvelines France Lebanese Univ Fac Engn CRSI LaRGES Beirut Lebanon

As the electrification of vehicles keeps being widespread, and facing the impossibility of storing big amounts of electrical energy, the challenge of controlling and adapting the electricity supply and demand has become a necessity. Therefore, electric vehicles could be an optimal solution for the storage and the retrieval of energy depending on the supply and demand of electricity. Thus, this study proposes an energetic strategy based on a multi-objective and multi-criteria optimization algorithm related to the control of the energy flows between the electric vehicles and the grid, home or building depending on the electricity supply and demand. The main focus at this stage involves the optimization of the vehicles' charging mode. Hence, in this paper, the multi-objective optimization proposed is first implemented through the presentation of its algorithm and the description and modeling of its objective functions and their corresponding constraints. Then, the genetic algorithm optimization approach is adopted for the charging process of the study to find the optimal solution for each objective function. Besides, the weighted sum approach is applied after the normalization of all the fitness functions and several case studies are carried out to find the optimal solutions based on the priority of specific objectives over the others. Once all calculations are done, a simulation via Matlab software is performed and the results are discussed and compared. The simulation of the results has verified the theoretical calculations proving the effectiveness of the proposed methodology. However, as some of the calculated optimal solutions of the system seem to be conflicting, the prioritization of some objectives over others had to be operated in order to figure out a global solution for the multi-objective system. Thus, as a compromise had to be applied for the calculation of quasi-optimal solutions that would converge towards the Pareto-front, the decision maker's preferenc

关键词： Electric vehicles Energy flows Multi-objective optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

An improved reversible watermarking scheme using embedding optimization and quaternion moments

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-29页

作者： Tahiri, Mohamed Amine Karmouni, Hicham Sayyouri, Mhamed Qjidaa, Hassan Ahmad, Musheer Hammad, Mohamed Plawiak, Pawel Alfarraj, Osama Abd El-Latif, Ahmed A. Sidi Mohamed Ben Abdellah Univ Natl Sch Appl Sci Engn Syst & Applicat Lab Fes Morocco Cadi Ayyad Univ Natl Sch Appl Sci Marrakech 40000 Morocco Sidi Mohamed Ben Abdellah Fez Univ Lab Elect Signals & Syst Informat Fes Morocco Jamia Millia Islamia Dept Comp Engn New Delhi 110025 India Menoufia Univ Fac Comp & Informat Dept Informat Technol Shibin Al Kawm 32511 Egypt Cracow Univ Technol Fac Comp Sci & Telecommun Dept Comp Sci Warszawska 24 PL-31155 Krakow Poland Polish Acad Sci Inst Theoret & Appl Informat Baltycka 5 PL-44100 Gliwice Poland King Saud Univ Community Coll Comp Sci Dept Riyadh 11437 Saudi Arabia Jadara Univ Res Ctr Irbid Jordan Menoufia Univ Fac Sci Dept Math & Comp Sci Shibin Al Kawm 32511 Egypt

Digital watermarking of images is an essential method for copyright protection and image security. This paper presents an innovative, robust watermarking system for color images based on moment and wavelet transformations, algebraic decompositions, and chaotic systems. First, we extended classical Charlier moments to quaternary Charlier moments (QCM) using quaternion algebra. This approach eliminates the need to decompose color images before applying the discrete wavelet transform (DWT), reducing the computational load. Next, we decompose the resulting DWT matrix using QR and singular value decomposition (SVD). To enhance the system's security and robustness, we introduce a modified version of Henon's 2D chaotic map. Finally, we integrate the arithmetic optimization algorithm to ensure dynamic and adaptive watermark insertion. Our experimental results demonstrate that our approach outperforms current color image watermarking methods in security, storage capacity, and resistance to various attacks, while maintaining a high level of invisibility.

关键词： Henon chaotic map Quaternion moments Digital image watermarking optimization algorithm

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：