检索结果-内蒙古大学图书馆

Design of audio to image cross-modal learning and generation based on single-layer CoPt spin-orbit torque devices

学校读者我要写书评

暂无评论

Journal of Magnetism and Magnetic Materials 2025年 629卷

作者： Likun Qian Liu Yang Chao Zuo Ying Tao Wendi Li Fang Jin Huihui Li Kaifeng Dong School of Future Technology China University of Geosciences Wuhan 430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan 430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan 430074 China School of Automation China University of Geosciences Wuhan 430074 China Wuhan Second Ship Design and Research Institute Wuhan 430064 China Beijing Superstring Academy of Memory Technology Beijing 100176 China

Despite remarkable achievements in neural network applications using spin–orbit torque (SOT) devices, current research predominantly focuses on unimodal classification tasks, while cross-modal learning tasks often suffer from suboptimal generation quality. To address this limitation, we employ generative adversarial networks (GANs) based on CoPt-SOT to achieve cross-modal learning and generation from speech to handwritten digit images. Initially, we explore the field-free switching characteristics of single-layer CoPt devices, developing spin-based implementations of Scaled Rectified Linear Unit (SReLu), Sigmoid, and Tanh neuronal functions. Simultaneously, we leverage the nonvolatile multistate characteristics of CoPt devices to realize excitatory-inhibitory synaptic plasticity. Subsequently, for the first time, we construct half-spin and full-spin GAN networks using these spin-based neurons and synapses, enabling cross-modal learning and generation between speech and image domains. Finally, we evaluate the performance of these networks through handwritten digit recognition tasks, achieving impressive recognition accuracies of 93.78 % and 88.61 % for half-spin and full-spin implementations, respectively. Notably, this work overcomes the existing bottleneck in SOT device applications, which have been largely confined to unimodal classification tasks, and demonstrates significant potential for expanding the scope of SOT-based technologies.

关键词： CoPt single-layer Cross-modal learning Field-free switching Neural network

Fully linear graph convolutional networks for semi-supervised learning and clustering

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Cai, Yaoming Zhang, Zijia Cai, Zhihua Liu, Xiaobo Ding, Yao Ghamisi, Pedram School of Computer Science China University of Geosciences Wuhan430074 China School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems China University of Geosciences Wuhan430074 China Xi'an Research Institute of High Technology Xi'an710000 China Helmholtz Institute Freiberg for Resource Technology Freiberg09599 Germany Vienna1030 Austria

This paper presents FLGC, a simple yet effective fully linear graph convolutional network for semi-supervised and unsupervised learning. Instead of using gradient descent, we train FLGC based on computing a global optimal closedform solution with a decoupled procedure, resulting in a generalized linear framework and making it easier to implement, train, and apply. We show that (1) FLGC is powerful to deal with both graph-structured data and regular data, (2) training graph convolutional models with closed-form solutions improve computational efficiency without degrading performance, and (3) FLGC acts as a natural generalization of classic linear models in the non-Euclidean domain, e.g., ridge regression and subspace clustering. Furthermore, we implement a semisupervised FLGC and an unsupervised FLGC by introducing an initial residual strategy, enabling FLGC to aggregate long-range neighborhoods and alleviate over-smoothing. We compare our semi-supervised and unsupervised FLGCs against many state-of-the-art methods on a variety of classification and clustering benchmarks, demonstrating that the proposed FLGC models consistently outperform previous methods in terms of accuracy, robustness, and learning efficiency. The core code of our FLGC is released at https://***/AngryCai/FLGC. Copyright © 2021, The Authors. All rights reserved.

关键词： Clustering algorithms

A Bioinspired retinal neural network for accurately extracting small-target motion information in cluttered backgrounds

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, Xiao Qiao, Hong Li, Hui Jiang, Zhihong School of Mechatronical Engineering Advanced Innovation Center for Intelligent Robots and Systems Key Laboratory of Biomimetic Robots and Systems of Chinese Ministry of Education Beijing Institute of Technology Beijing100081 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Beijing Key Laboratory of Research and Application for Robotic Intelligence of’Hand-Eye-Brain’ Interaction Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

—Robust and accurate detection of small moving targets in cluttered moving backgrounds is a significant and challenging problem for robotic visual systems to perform search and tracking tasks. Inspired by the neural circuitry of elementary motion vision in the mammalian retina, this paper proposes a bioinspired retinal neural network based on a new neurodynamics-based temporal filtering and multiform 2-D spatial Gabor filtering. This model can estimate motion direction accurately via only two perpendicular spatiotemporal filtering signals, and respond to small targets of different sizes and velocities through adjusting the dendrite field size of spatial filter. Meanwhile, an algorithm of directionally selective inhibition is proposed to suppress the target-like features in the moving background, which can reduce the influence of background motion effectively. Extensive synthetic and real-data experiments show that the proposed model works stably for small targets of a wider size and velocity range, and has better detection performance than other bioinspired models. Additionally, it can also extract the information of motion direction and motion energy accurately and rapidly. Copyright © 2021, The Authors. All rights reserved.

关键词： Ophthalmology

A Bioinspired approach-sensitive neural network for collision detection in cluttered and dynamic backgrounds

学校读者我要写书评

暂无评论

arXiv 2021年

—Rapid, accurate and robust detection of looming objects in cluttered moving backgrounds is a significant and challenging problem for robotic visual systems to perform collision detection and avoidance tasks. Inspired by the neural circuit of elementary motion vision in the mammalian retina, this paper proposes a bioinspired approach-sensitive neural network (ASNN) that contains three main contributions. Firstly, a direction-selective visual processing module is built based on the spatiotemporal energy framework, which can estimate motion direction accurately via only two mutually perpendicular spatiotemporal filtering channels. Secondly, a novel approach-sensitive neural network is modeled as a push-pull structure formed by ON and OFF pathways, which responds strongly to approaching motion while insensitivity to lateral motion. Finally, a method of directionally selective inhibition is introduced, which is able to suppress the translational backgrounds effectively. Extensive synthetic and real robotic experiments show that the proposed model is able to not only detect collision accurately and robustly in cluttered and dynamic backgrounds but also extract more collision information like position and direction, for guiding rapid decision making. © 2021, CC BY.

关键词： Mammals

Adaptive Fuzzy control for Fractional-Order Interconnected systems with Unknown control Directions

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Liang, Bingyun Zheng, Shiqi Ahn, Choon Ki School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China School of Electrical Engineering Korea University Seoul136-701 Korea Republic of

This paper concentrates on the study of the decentralized fuzzy control method for a class of fractional-order interconnected systems with unknown control directions. To overcome the difficulties caused by the multiple unknown control directions in fractional-order systems, a novel fractional-order Nussbaum function technique is proposed. This technique is much more general than those of existing works since it not only handles single/multiple unknown control directions but is also suitable for fractional/integer-order single/interconnected systems. Based on this technique, a new decentralized adaptive control method is proposed for fractional-order interconnected systems. Smooth functions are introduced to compensate for unknown interactions among subsystems adaptively. Furthermore, fuzzy logic systems are utilized to approximate unknown nonlinearities. It is proven that the designed controller can guarantee the boundedness of all signals in interconnected systems and the convergence of tracking errors. Two examples are given to show the validity of the proposed method. Copyright © 2020, The Authors. All rights reserved.

关键词： Algebra

UAV Image Mosaicing Based Multi-Region Local Projection Deformation

学校读者我要写书评

暂无评论

UAV Image Mosaicing Based Multi-Region Local Projection Defo...

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Quan Xu Linbo Luo Jun Chen Wenping Gong Donghai Guo School of Mechanical Engineering and Electronic Information China University of Geosciences Wuhan School of Automation China University of Geosciences Wuhan Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Faculty of Engineering China University of Geosciences Wuhan

ISBN: (数字)9781728163741

ISBN: (纸本)9781728163758

The goal of unmanned aerial vehicle (UAV) image mosaicing is to create natural-looking mosaics free of artifacts due to the parallax of the image and relative camera motion. UAV remote sensing is a low-altitude technology and the UAV imaged scene is not effectively planar, yielding parallax on images. In this paper, we apply local homography to match UAV images, which can reduce misalignment artifacts or “ghosting” in the results compared with 2D projective transforms or global homography. In addition, when an object in three dimensions is mapped to an image plane, different surfaces have different projections. These projections vary with the viewpoint in a sequence of UAV images, which still causes artifacts near some tall buildings if we only use local homography. We propose a novel stitching method based multi-region local projection deformation, that divides the overlapping regions of input images into several regions, then meshes image to calculate local projections by partitioned regions. Specifically, we use a strategy where multiple regions have different weights for calculating local projections, which can significantly reduce ghosting due to these projections vary with the viewpoint and parallax. The benefits of the proposed approach are demonstrated using a variety of challenging cases.

关键词： Unmanned aerial vehicles Remote sensing Geology Two dimensional displays Task analysis Buildings Transmission line matrix methods

nonlocal quark condensate from Dyson-Schwinger Equation and its contributions to the gluon vacuum polarization based on OPE approach

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, Jing-Hui Duan, Xue-Ying Huan, Chen Wang, Guang-Jun Hu, Xiang-Yun School of Institute of Geophysics and Geomatics China University of Geosciences Lumo Road 388 Wuhan430074 China School of Automation China University of Geosciences Lumo Road 388 Wuhan430074 China School of Mathematics and Physics China University of Geosiciences Lumo Road 388 Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education China

The operator-product expansion(OPE) could be employed to obtain the lowest-order, nonlocal quark scalar condensate component of gluon vacuum polarization. In particular, nonlocal quark scalar condensate can be calculated by solving Dyson-Schwinger Equation(DSE) of QCD. Then, field-theoretic aspects of the gluon vacuum polarization and nonperturbative gluon propagator will be considered in the Landau gauge of the Lorentz gauge fixing. The gluon propagator we obtained is finite in the infrared domain where the single gluon mass mg can be determined. Our results of the ratio mg/ΛQCD the range of that from 1.33 to 1.39 agree with previous determinations for this ratio. Besides, the analytic structure of the gluon propagators from the OPE's result is explored. Our numerical analysis of the gluon' Schwinger function finds clear evidence of the positivity violations in the gluon propagator. In addition, a new method for obtaining the chemical potential dependence of the gluon vacuum polarization and the dressed gluon propagator is developed. © 2021, CC BY.

关键词： Polarization

Design of Smart Home Monitoring and control System Based on Zigbee and WIFI

学校读者我要写书评

暂无评论

Design of Smart Home Monitoring and Control System Based on ...

第三十八届中国控制会议

作者： Chentao Li Kaifeng Dong Fang Jin Junlei Song Wenqin Mo School of Automation China University of Geosciences Hubei key Laboratory of Advanced Control and Intelligent Automation for Complex Systems

Smart Home system is the use of home life associated with hardware and related application technology, the introduction of home appliance automation, remote control, intelligent processing technology in a traditional home environment, to design a provide users with comprehensive information exchange capabilities, in order to enhanced home life safety, comfort and convenience of home systems. At present, there are problems such as complicated wiring, single service, and high price in this field. In response to these problems, this paper designs a smart home system based on Zigbee and WIFI. The system consists of a terminal node, a gateway node and a remote control terminal. The terminal node and the gateway node form a home internal network through Zigbee technology, and the remote control terminal accesses and controls the home internal network through WIFI.

关键词： Smart home Zigbee WIFI Wireless transmission Remote control

control System Design for Weight on Bit in complex Geological Drilling Process

学校读者我要写书评

暂无评论

Control System Design for Weight on Bit in Complex Geologica...

第三十八届中国控制会议

作者： Sike Ma Min Wu Jun Cheng Luefeng Chen School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems

With the increase in drilling depth, the encountered formations become more complicated, and industrial controller for weight on bit has limited precision due to the uncertain and varying formations. In this paper, we design a scheme of control system for weight on bit to achieve higher control precision in complex geological drilling process. A double-loop control structure is introduced, which involves with two system models: a multi-degree-of-freedom lumped parameter drill string model and a finite element bit-rock interaction model. The effect of varying formations on weight on bit is analyzed, and a control strategy for control system is presented in this paper.

关键词： Drilling process weight on bit control system drill string model bit-rock interaction model uncertain formations