检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Leitner, Stefan Mirza, M. Jehanzeb Lin, Wei Micorek, Jakub Masana, Marc Kozinski, Mateusz Possegger, Horst Bischof, Horst Institute for Computer Graphics and Vision Graz University of Technology Austria Christian Doppler Laboratory for Embedded Machine Learning Austria Christian Doppler Laboratory for Semantic 3D Computer Vision Austria TU Graz SAL Dependable Embedded Systems Lab Silicon Austria Labs Austria

In autonomous driving scenarios, current object detection models show strong performance when tested in clear weather. However, their performance deteriorates significantly when tested in degrading weather conditions. In addition, even when adapted to perform robustly in a sequence of different weather conditions, they are often unable to perform well in all of them and suffer from catastrophic forgetting. To efficiently mitigate forgetting, we propose Domain-Incremental Learning through Activation Matching (DILAM), which employs unsupervised feature alignment to adapt only the affine parameters of a clear weather pre-trained network to different weather conditions. We propose to store these affine parameters as a memory bank for each weather condition and plug-in their weather-specific parameters during driving (i.e. test time) when the respective weather conditions are encountered. Our memory bank is extremely lightweight, since affine parameters account for less than 2% of a typical object detector. Furthermore, contrary to previous domain-incremental learning approaches, we do not require the weather label when testing and propose to automatically infer the weather condition by a majority voting linear classifier. © 2023, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Scalable and low-cost remote lab platforms: Teaching industrial robotics using open-source tools and understanding its social implications

arXiv

引用

arXiv 2024年

作者： Kumar, Amit Jose, Jaison Jain, Archit Kulkarni, Siddharth Arya, Kavi Centre for Systems and Control Indian Institute of Technology Bombay Mumbai India Embedded Real-Time Systems / e-Yantra Lab. Indian Institute of Technology Bombay Mumbai India Computer Science and Engineering Department Indian Institute of Technology Bombay Mumbai India

With recent advancements in industrial robots, educating students in new technologies and preparing them for the future is imperative. However, access to industrial robots for teaching poses challenges, such as the high cost of acquiring these robots, the safety of the operator and the robot, and complicated training material. This paper proposes two low-cost platforms built using open-source tools like Robot Operating System (ROS) and its latest version ROS 2 to help students learn and test algorithms on remotely connected industrial robots. Universal Robotics (UR5) arm and a custom mobile rover were deployed in different life-size testbeds, a greenhouse, and a warehouse to create an Autonomous Agricultural Harvester System (AAHS) and an Autonomous Warehouse Management System (AWMS). These platforms were deployed for a period of 7 months and were tested for their efficacy with 1,433 and 1,312 students, respectively. The hardware used in AAHS and AWMS was controlled remotely for 160 and 355 hours, respectively, by students over a period of 3 months. © 2024, CC BY.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

Training Adversarially Robust Sparse Networks via Bayesian Connectivity Sampling 38

Training Adversarially Robust Sparse Networks via Bayesian C...

引用

38th International Conference on Machine Learning, ICML 2021

作者： Özdenizci, Ozan Legenstein, Robert Graz University of Technology Institute of Theoretical Computer Science Graz Austria Silicon Austria Labs TU Graz - SAL Dependable Embedded Systems Lab Graz Austria

ISBN: (纸本)9781713845065

Deep neural networks have been shown to be susceptible to adversarial attacks. This lack of adversarial robustness is even more pronounced when models are compressed in order to meet hardware limitations. Hence, if adversarial robustness is an issue, training of sparsely connected networks necessitates considering adversarially robust sparse learning. Motivated by the efficient and stable computational function of the brain in the presence of a highly dynamic synaptic connectivity structure, we propose an intrinsically sparse rewiring approach to train neural networks with state-of-the-art robust learning objectives under high sparsity. Importantly, in contrast to previously proposed pruning techniques, our approach satisfies global connectivity constraints throughout robust optimization, i.e., it does not require dense pre-training followed by pruning. Based on a Bayesian posterior sampling principle, a network rewiring process simultaneously learns the sparse connectivity structure and the robustness-accuracy trade-off based on the adversarial learning objective. Although our networks are sparsely connected throughout the whole training process, our experimental benchmark evaluations show that their performance is superior to recently proposed robustness-aware network pruning methods which start from densely connected networks. Copyright © 2021 by the author(s)

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

On transitioning from PID to ADRC in thermal power plants

引用

Control Theory and Technology 2021年第1期19卷 3-18页

作者： Zhenlong Wu Zhiqiang Gao Donghai Li YangQuan Chen Yanhong Liu School of Electrical Engineering Zhengzhou UniversityZhengzhou 450001HenanChina Department of Electrical and Computer Engineering Cleveland State UniversityClevelandOH 44115USA State Key Lab of Power Systems Department of Energy and Power EngineeringTsinghua UniversityBeijing 100084China Mechatronics Embedded Systems and Automation(MESA)LabDepartment of Mechanical EngineeringSchool of EngineeringUniversity of CaliforniaMercedCA 95343USA

This paper focuscs on the recent progress in the adoption of active disturbance rejection control(ADRC)in thermal pro-cesses as a viable alternative to proportional-_integral-derivative(PID),especially in coa-fired power *** profound interpretation of this paradigm shift,with backward compatibility,is discussed in detail.A few fundamental issues associated with ADRC's applications in thermal processes are discussed,such as implementation,tuning,and the structural *** and case studies are presented,encompassing coal-fired power plants,gas turbines and nuclear power plants,as well as highlighting results of field *** discussed are future research opportunities brought by ADRC's entry as the baseline control technology in thermal processes.

关键词： Active disturbance rejection control(ADRC) Proportional-integral-derivative(PID) Thermal processes Coal-fired power plant Field applications

来源：评论

学校读者我要写书评

暂无评论

Quantum Calculus-Based Two-Dimensional Least Mean Square Algorithm

Quantum Calculus-Based Two-Dimensional Least Mean Square Alg...

引用

Asilomar Conference on Signals, systems & computers

作者： Muhammad Moinuddin Azzedine Zerguine Ubaid M. Al-Saggaf Muhammad Arif Electrical and Computer Engineering Department and the Centre of Excellence in Intelligent Engineering Systems King Abdulaziz University Jeddah Saudi Arabia Electrical Engineering Department and the Interdisciplinary Research Center for Communication Systems and Sensing KFUPM Dhahran Saudi Arabia Institute of Networked and Embedded Systems University of Klagenfurt and the Ubiquitous Sensing Systems Lab University of Klagenfurt-Silicon Austria Labs Klagenfurt Austria

ISBN: (数字)9798350354058

ISBN: (纸本)9798350354065

Recent work of the one-dimensional q-LMS algorithm give insight that the Quantum Calculus based gradient called the q-gradient is capable of improving the convergence behavior in contrast to the standard LMS algorithm. The reason for this is the fact that the q-derivative takes larger steps in minimizing a cost function as it evaluates the secant in contrast to the tangent of the cost function. Motivated by this, we propose a two-dimensional version of the q-LMS algorithm. More precisely, we employ Quantum Calculus based q-gradient in steepest descent optimization of mean-square-error (MSE) cost function in 2D adaptive filtering scenario to obtain two-dimensional q-LMS (2D q-LMS) algorithm. A thorough analytical investigation of the proposed 2D q-LMS algorithm in both mean and mean-square sense is provided. Both the transient and steady-state convergence behaviours are examined. Consequently, excess MSE (EMSE) learning curve and its steady-state expression are evaluated in closed form. Simulation results are presented for the application of noise cancelation in images to show the superiority of the proposed 2D q-LMS algorithm.

关键词： Filtering Two-dimensional displays Simulation Cost function Calculus Steady-state Performance analysis Transient analysis Standards Convergence

来源：评论

学校读者我要写书评

暂无评论

Optical Character Recognition Using Optimized Convolutional Networks*

Optical Character Recognition Using Optimized Convolutional ...

引用

International Conference on Fog and Mobile Edge Computing (FMEC)

作者： Anum Nawaz Muhammad Irfan Tomi Westerlund Shanghai Key Laboratory of Intelligent Information Processing School of Computer ScienceFudan University Shanghai China Turku Intelligent and Embedded Robotic Systems Lab Faculty of Technology University of Turku Turku Finland

Optical Character Recognition (OCR) has been a prominent area of research in pattern recognition for several decades, owing to its broad application potential in smart living. To improve offline OCR on mobile devices with limited computing resource, we have optimized Convolutional Neural Networks (CNNs) to efficiently detect text using minimal resources. To achieve this, we employed two distinct pretrained CNN models, namely AlexNet and Inception-V3, for feature extraction. Leveraging these models' unique characteristics and capabilities to extract diverse features, we aimed to enhance the classifier's accuracy. This, in turn, facilitates the development of an efficient edge-device application for faster and higher-quality OCR. Experimental results demonstrate that our proposed optimized algorithm outperforms existing CNN-based methods in the field of OCR, particularly in the categorization and detection of handwritten digits and character recognition. The conducted research yielded impressive accuracy results, with up to 97% accuracy on the MNIST dataset and 95.5% accuracy on the NIST dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Efficient Domain-Incremental Learning Approach to Drive in All Weather Conditions

arXiv

引用

arXiv 2022年

作者： Mirza, Muhammad Jehanzeb Masana, Marc Possegger, Horst Bischof, Horst Institute of Computer Graphics and Vision Graz University of Technology Austria Christian Doppler Laboratory for Embedded Machine Learning Austria TU Graz SAL Dependable Embedded Systems Lab Silicon Austria Labs Austria

Although deep neural networks enable impressive visual perception performance for autonomous driving, their robustness to varying weather conditions still requires attention. When adapting these models for changed environments, such as different weather conditions, they are prone to forgetting previously learned information. This catastrophic forgetting is typically addressed via incremental learning approaches which usually re-train the model by either keeping a memory bank of training samples or keeping a copy of the entire model or model parameters for each scenario. While these approaches show impressive results, they can be prone to scalability issues and their applicability for autonomous driving in all weather conditions has not been shown. In this paper we propose DISC - Domain Incremental through Statistical Correction - a simple online zero-forgetting approach which can incrementally learn new tasks (i.e. weather conditions) without requiring re-training or expensive memory banks. The only information we store for each task are the statistical parameters as we categorize each domain by the change in first and second order statistics. Thus, as each task arrives, we simply 'plug and play' the statistical vectors for the corresponding task into the model and it immediately starts to perform well on that task. We show the efficacy of our approach by testing it for object detection in a challenging domain-incremental autonomous driving scenario where we encounter different adverse weather conditions, such as heavy rain, fog, and snow. © 2022, CC BY.

关键词： Meteorology

来源：评论

学校读者我要写书评

暂无评论

STM32 miner : A lightweight blockchain mining system

STM32 miner : A lightweight blockchain mining system

引用

IEEE International Conference on Design & Test of Integrated Micro & Nano-systems (DTS)

作者： Jalel Ktari Hedi Choura Tarek Frikha Mouna Baklouti Dhia Mejdi Molka Kotti Ahmed Zribi Hassine Touil Ghaith Oueslati Computer Embedded Systems Lab National Engineering School of Sfax Sfax Tunisia Data Engineering and Semantics Unit National Engineering School of Sfax Sfax Tunisia Computer Sciences and Applied Mathematiques Departement National Engineering School of Sfax Sfax Tunisia

Despite being a promising area of exploration, the application of embedded systems in blockchain technology, particularly with FFGA or Raspberry technology, remains relatively limited. Despite the advantages offered by FPGAs, such as high-speed memory, parallel computing blocks, and a flexible architecture, their use in blockchain has primarily been focused on Proof of Work (PoW) consensus *** this paper, our goal is to address this gap by presenting the implementation of a blockchain application specifically tailored for the STM32 platform. By leveraging the capabilities of the STM32 microcontroller, we aim to demonstrate the practicality and effectiveness of embedded systems in blockchain technology, overcoming the limitations observed with previous platforms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Vision-based indoor localization of nano drones in controlled environment with its applications

arXiv

引用

arXiv 2024年

作者： Singh, Simranjeet Kumar, Amit Chemban, Fayyaz Pocker Fernandes, Vikrant Penubaku, Lohit Arya, Kavi Electrical Engineering Department Indian Institute of Technology Bombay Mumbai India Embedded Real-Time Systems e-Yantra Lab Indian Institute of Technology Bombay Mumbai India Computer Science and Engineering Department Indian Institute of Technology Bombay Mumbai India

Navigating unmanned aerial vehicles in environments where GPS signals are unavailable poses a compelling and intricate challenge. This challenge is further heightened when dealing with Nano Aerial Vehicles (NAVs) due to their compact size, payload restrictions, and computational capabilities. This paper proposes an approach for localization using off-board computing, an off-board monocular camera, and modified open-source algorithms. The proposed method uses three parallel proportional-integral-derivative controllers on the off-board computer to provide velocity corrections via wireless communication, stabilizing the NAV in a custom-controlled environment. Featuring a 3.1cm localization error and a modest setup cost of 50 USD, this approach proves optimal for environments where cost considerations are paramount. It is especially well-suited for applications like teaching drone control in academic institutions, where the specified error margin is deemed acceptable. Various applications are designed to validate the proposed technique, such as landing the NAV on a moving ground vehicle, path planning in a 3D space, and localizing multi-NAVs. The created package is openly available at https://***/simmubhangu/eyantra drone to foster research in this field. © 2024, CC BY.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence for securing SDIoT: challenges and countermeasures

Artificial Intelligence for securing SDIoT: challenges and c...

引用

IEEE SSD International Multi-Conference on systems, Signals and Devices

作者： Ameur Salem Zaidoun Hajer Baccouch Abderrahmen Guermazi Zied Lachiri Signal Images and Information Technologies Lab National Engineering School of Tunis Tunisia Higher Institute of Technological Studies of Siliana Tunisia Computer and Embedded Systems Lab National Engineering School of Sfax Tunisia Higher Institute of Technological Studies of Sfax Tunisia

SDIoT (Software Defined IoT) is an emerging IoT technology, deployed using SDN architecture; nevertheless, it presents several weaknesses, which can be raised in different layers. To mitigate SDIoT intrusions and attacks, security as used in a common network cannot provide a complete solution and must be enhanced with AI (Artificial Intelligence) based countermeasures tools, mainly machine and deep learning. The paper identifies the salient features of SDIoT technology. Afterwards, it describes the most common attacks against SDIoT networks and possible countermeasures using several deep learning algorithms. In addition, the paper describes datasets used for learning steps. At the end, it discusses the suitability of AI based security tools to some attacks.

关键词：

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：