检索结果-内蒙古大学图书馆

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Ionut M. Motoi Leonardo Saraceni Daniele Nardi Thomas A. Ciarfuglia Department of Computer Control and Management Engineering Sapienza University of Rome Rome Italy

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Satellite imagery is crucial for tasks like environmental monitoring and urban planning. Typically, it relies on semantic segmentation or Land Use Land Cover (LULC) classification to categorize each pixel. Despite the advancements brought about by Deep Neural Networks (DNNs), their performance in segmentation tasks is hindered by challenges such as limited availability of labeled data, class imbalance and the inherent variability and complexity of satellite images. In order to mitigate those issues, our study explores the effectiveness of a Cut-and-Paste augmentation technique for semantic segmentation in satellite images. We adapt this augmentation, which usually requires labeled instances, to the case of semantic segmentation. By leveraging the connected components in the semantic segmentation labels, we extract instances that are then randomly pasted during training. Using the DynamicEarthNet dataset and a U-Net model for evaluation, we found that this augmentation significantly enhances the mIoU score on the test set from 37.9 to 44.1. This finding highlights the potential of the Cut-and-Paste augmentation to improve the generalization capabilities of semantic segmentation models in satellite imagery.

关键词： Training Adaptation models Semantic segmentation Semantics Urban planning Land surface Satellite images

来源：评论

学校读者我要写书评

暂无评论

Policy for Optimal Avoidance of Multiple Engagement Zones

Policy for Optimal Avoidance of Multiple Engagement Zones

引用

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

作者： Milutinović, Dejan Von Moll, Alexander Weintraub, Isaac E. Casbeer, David Department of Electrical and Computer Engineering University of California Santa Cruz Santa CruzCA95064 United States Control Science Center Air Force Research Laboratory WPAFB OH45433 United States

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

This work formulates the feedback control strategies for vehicles to reach a goal point amongst a field of dynamic risk regions. Whereas previous work has considered deterministic versions of this problem, we consider the scenario in the context of uncertainty. This uncertainty could be due to unknown wind or other external disturbances. The risk regions’ dynamics are tightly coupled to the state of the vehicle and as a result the task of navigating through a field dynamically changes as the vehicle traverses through it. Rather than using a nonlinear program solver, the approach taken here is one of formulating the problem as a stochastic optimal control problem wherein a feedback control law is derived. This feedback control law allows a single or multiple vehicles to reach the desired goal location efficiently and with minimal risk incursion. When inside a risk region, there is a probability rate of the vehicle being neutralized. The stochastic dynamical system is modeled as a hybrid system, i.e., regime switching diffusion. The hybrid system is comprised of continuous spatial dynamics and discrete operational states corresponding to "normal" and "neutralized". The optimal control problem is discretized into a field of vehicle states and control. Using Value Iteration, the control signal at each discrete location is optimized, providing a feedback control strategy over the field. Following a discussion of the methods and approaches, this problem is formulated, simulations are performed, and conclusions and remarks are made © 2025, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Tabular Dyna-Q Algorithm for Online Calculation of LQR Gains: A Simulation Study 9

Tabular Dyna-Q Algorithm for Online Calculation of LQR Gains...

引用

9th International Conference on engineering and Emerging Technology, ICEET 2023

作者： Farooghi, Kian Samadi, Mahdieh Khaloozadeh, Hamid K. N. Toosi University of Technology Department of Systems and Control Tehran Iran University of Tabriz Faculty of Electrical and Computer Engineering Tabriz Iran

ISBN: (纸本)9798350316926

This paper presents a novel approach for the online calculation of Linear Quadratic Regulator (LQR) gains using the Tabular Dyna-Q algorithm. By leveraging Q-learning, this technique enables the determination of gains for unknown dynamic systems by using input-output data only. The proposed algorithm is particularly advantageous for situations where acquiring input-output data is expensive or time-consuming, as it effectively reduces computation time by reusing previous data. To validate the accuracy of the calculations and control scheme, a simulation example is provided. The results indicate the effectiveness of the Tabular Dyna-Q algorithm in accurately computing LQR gains and its potential application in real-world scenarios. This research contributes to advancing the field of control theory by offering an efficient and reliable method for online gain calculation in dynamic systems. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning Scheduling Research for Edge Servers 6

Reinforcement Learning Scheduling Research for Edge Servers

引用

6th International Conference on computer Network, Electronic and Automation, ICCNEA 2023

作者： Han, Yonggui Wang, Zhongsheng Xi'an Technological University School of Computer Science and Engineering Xi'an710021 China Monitoring and Control xi'An Technological University State and Provincial Joint Engineering Lab. of Advanced Network Xi'an China

ISBN: (纸本)9798350305388

The rapid development of the Internet of Things and big data has brought about massive amounts of data and the need for real-time responses, which traditional computing models can no longer meet. Edge computing comes into being, which pushes computing resources and services closer to the source of data generation so that data can be processed and analyzed closer to the user. The scheduling of siting strategies for edge servers is crucial for improving the performance of edge computing networks. Traditional scheduling methods for siting strategies are often based on rules or optimization algorithms, which lack adaptive and intelligent management of the network *** paper proposes a reinforcement learning DQN-based edge server siting policy scheduling method, which aims to achieve intelligent management and resource optimization of edge computing networks by learning the optimal policy. Experimental results show that the method has significant performance advantages in terms of delay, load balancing and energy consumption. The experiment adopts DQN algorithm, which is based on Q-learning and guides the agent to make decisions by optimizing Q-value function. In this paper, the reinforcement learning scheduling algorithm of edge server is studied and compared with other three algorithms. According to the experimental results, the proposed reinforcement learning scheduling algorithm reduces the delay by 7%, the energy consumption by 3%, and the load balancing by 20% compared with the other three algorithms. This shows that the algorithm has significant advantages in improving system performance. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Influence of Different Optimization Aalgorithms on Prediction Accuracy of Photovoltaic Output Power Based on BP Neural Network 41

Influence of Different Optimization Aalgorithms on Predictio...

引用

第41届中国控制会议

作者： Minglu Xu Chao Ma Xiaojuan Han Beijing Changping Power Supply Company North China Electric Power University School of Control Science and Computer Engineering Beijing

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

Aiming at the problem that back propagation(BP)neural network is easy to fall into local extremum,based on the short-term prediction of photovoltaic output power using the BP neural network,this paper studies the influence of genetic algorithm(GA) and particle swarm optimization(PSO) algorithm respectively to optimize the structure of the BP neural network on the prediction accuracy of photovoltaic output *** effectiveness of the proposed method is verified by the simulation of the actual operation data of a photovoltaic power *** simulation results show that different optimization algorithms have a significant impact on the prediction accuracy of photovoltaic output power.

关键词： photovoltaic output power prediction particle swarm optimization back propagation neural network

来源：评论

学校读者我要写书评

暂无评论

Biological Influence of Low-Frequency Electromagnetic Field Exposition Based on Selected Cell Death Markers 15

Biological Influence of Low-Frequency Electromagnetic Field ...

引用

15th International Conference Elektro, ELEKTRO 2024

作者： Cichorowski, Robert Kocikowska, Olga Bajtos, Marek Radil, Roman Janousek, Ladislav De Carrillo, Daria Gendosz Student, Sebastian Silesian University of Technology Faculty of Automatic Control Electronics and Computer Science Gliwice Poland Faculty of Automatic Control Electronics and Computer Science Department of Engineering and Systems Biology Gliwice Poland Silesian University of Medicine Faculty of Medical Sciences Department of Physiology Katowice Poland Faculty of Electrical Engineering and Informational Technology Department of Electromagnetic and Biomedical Engineering Žilina Slovakia Silesian University of Medicine Faculty of Medical Sciences in Zabrze Department of Histology Zabrze Poland Silesian University of Technology Biotechnology Center Gliwice Poland

ISBN: (纸本)9798350372359

This work presents the results of the examination of the HeLa cell line exposure on the ELF-EMF (extremely low-frequency electromagnetic field). In particular, the relationship between ELF-EMF exposition time and cell death. The examination of cell death hallmarks were examined by estimation levels of selected proteins - FACL4 (a protein that is a part of the ferroptosis pathway) and CK18 (Cytokeratin related to necrosis and apoptosis pathways) in the proposed model of workers exposed to ELF-EMF week. © 2024 IEEE.

关键词： Electromagnetic fields

来源：评论

学校读者我要写书评

暂无评论

Spiker+: a framework for the generation of efficient Spiking Neural Networks FPGA accelerators for inference at the edge

arXiv

引用

arXiv 2024年

作者： Carpegna, Alessio Savino, Alessandro Di Carlo, Stefano Department of Control and Computer Engineering of Politecnico di Torino Torino10129 Italy

Including Artificial Neural Networks (ANNs) in embedded systems at the edge allows applications to exploit Artificial Intelligence (AI) capabilities directly within devices operating at the network periphery, facilitating real-time decision-making. Especially critical in domains such as autonomous vehicles, industrial automation, and healthcare, the use of ANNs can enable these systems to process substantial data volumes locally, thereby reducing latency and power consumption. Moreover, it enhances privacy and security by containing sensitive data within the confines of the edge device. The adoption of Spiking Neural Networks (SNNs) in these environments offers a promising computing paradigm, mimicking the behavior of biological neurons and efficiently handling dynamic, time-sensitive data. However, deploying efficient SNNs in resource-constrained edge environments requires hardware accelerators, such as solutions based on Field Programmable Gate Arrays (FPGAs), that provide high parallelism and reconfigurability. This paper introduces Spiker+, a comprehensive framework for generating efficient, low-power, and low-area customized SNNs accelerators on FPGAs for inference at the edge. Spiker+ presents a configurable multi-layer hardware SNNs, a library of highly efficient neuron architectures, and a design framework, enabling the development of complex neural network accelerators with few lines of Python code. Spiker+ is tested on two benchmark datasets, the MNIST and the Spiking Heidelberg Dataset (SHD). On the MNIST, it demonstrates competitive performance compared to state-of-the-art SNN accelerators. It outperforms them in terms of resource allocation, with a requirement of 7,612 logic cells and 18 Block RAMs (BRAMs), which makes it fit in very small FPGAs, and power consumption, draining only 180mW for a complete inference on an input image. The latency is comparable to the ones observed in the state-of-the-art, with 780μs/img. To the authors' knowledge, Spiker+

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimizing DNN Inference on Multi-Accelerator SoCs at Training-time

arXiv

引用

arXiv 2024年

作者： Risso, Matteo Burrello, Alessio Pagliari, Daniele Jahier Department of Control and Computer Engineering Politecnico di Torino Turin10129 Italy

The demand for executing Deep Neural Networks (DNNs) with low latency and minimal power consumption at the edge has led to the development of advanced heterogeneous Systems-on-Chips (SoCs) that incorporate multiple specialized computing units (CUs), such as accelerators. Offloading DNN computations to a specific CU from the available set often exposes accuracy vs efficiency trade-offs, due to differences in their supported operations (e.g., standard vs. depthwise convolution) or data representations (e.g., more/less aggressively quantized). A challenging yet unresolved issue is how to map a DNN onto these multi-CU systems to maximally exploit the parallelization possibilities while taking accuracy into account. To address this problem, we present ODiMO, a hardware-aware tool that efficiently explores fine-grain mapping of DNNs among various on-chip CUs, during the training phase. ODiMO strategically splits individual layers of the neural network and executes them in parallel on the multiple available CUs, aiming to balance the total inference energy consumption or latency with the resulting accuracy, impacted by the unique features of the different hardware units. We test our approach on CIFAR-10, CIFAR-100, and ImageNet, targeting two open-source heterogeneous SoCs, i.e., DIANA and Darkside. We obtain a rich collection of Pareto-optimal networks in the accuracy vs. energy or latency space. We show that ODiMO reduces the latency of a DNN executed on the Darkside SoC by up to 8× at iso-accuracy, compared to manual heuristic mappings. When targeting energy, on the same SoC, ODiMO produced up to 50.8× more efficient mappings, with minimal accuracy drop ( © 2024, CC BY.

关键词： System-on-chip

来源：评论

学校读者我要写书评

暂无评论

DV-Hop Localization Strategy Based on Beacon Node Confidence Analysis 6

DV-Hop Localization Strategy Based on Beacon Node Confidence...

引用

6th International Conference on computer Network, Electronic and Automation, ICCNEA 2023

作者： Shen, Han Wang, Zhongsheng Xi'an Technological University School of Computer Science and Engineering Xi'an710021 China Xi'an Technological University State and Provincial Joint Engineering Lab. of Advanced Network Monitoring and Control Xi'an China

ISBN: (纸本)9798350305388

To address the problem that the DV-Hop localization model for wireless sensor networks can no longer meet the current localization accuracy requirements, this paper proposes the concept of beacon node trustworthiness to provide ideas and basis for improving the optimization model. Based on the principle of least squares in the localization model, weighted least squares is introduced to present the trustworthiness for beacon nodes with different characteristics. To analyze the characteristics of the trustworthiness of beacon nodes, 18 functions are designed in terms of function instrumentality and decrement able, hierarchical faultiness, etc. Finally, the conclusions obtained from the credibility test are tested with weighted matrix and average hop distance weighting, and the localization error is reduced from 32.68% to 31.78%, with a role rate of 100%. The credibility characteristics of beacon nodes discussed in this paper provide ideas for subsequent optimization of the localization model. © 2023 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Process-Model-Free Design of MPC via a Reference-Model-Based PI controller for the Gas Turbine 9

Process-Model-Free Design of MPC via a Reference-Model-Based...

引用

9th International Conference on Electrical engineering, control and Robotics, EECR 2023

作者： Shi, Gengjin Dai, Chengbo Ma, Miaomiao Li, Donghai Ding, Yanjun Tsinghua University State Key Lab of Power Systems Dept. of Energy & Power Engineering Beijing China North China Electric Power University School of Control & Computer Engineering Beijing China

ISBN: (纸本)9781665491204

Model predictive control (MPC) is a modern advanced control strategy which has great reputation because of its excellent reference tracking performance and the ability to deal with process constraints, time delay and multivariable problems. However, the gas turbine is difficult to model accurately and unmeasurable disturbances are unavoidable, which limits the application of MPC to the gas turbine. To solve these problems, a process-model-free design method for MPC via a reference-model (RM)-based proportional-integral (PI) controller-desired dynamic equational (DDE)-PI controller (MPC-DDE) is proposed in this paper. The DDE-PI provides the design model for the MPC and handle with unmeasurable disturbances while MPC provides a rolling optimized set point for DDE-PI. Simulation results illustrate the superiorities of the proposed hierarchical control strategy in reference tracking, disturbance rejection and robustness, which indicates the promising prospect of practical application of MPC-DDE to the gas turbine. © 2023 IEEE.

关键词： Model predictive 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：