Learning knowledge representation with meta knowledge distillation for single image super-resolution

Although the deep CNN-based super-resolution methods have achieved outstanding performance, their memory cost and computational complexity severely limit their practical employment. Knowledge distillation (KD), which can efficiently transfer knowledge from a cumbersome network (teacher) to a compact network (student), has demonstrated its advantages in some computer vision applications. The representation of knowledge is vital for knowledge transferring and student learning, which is generally defined in hand-crafted manners or uses the intermediate features directly. In this paper, we propose a model-agnostic meta knowledge distillation method under the teacher–student architecture for the single image super-resolution task. It provides a more flexible and accurate way to help teachers transmit knowledge in accordance with the abilities of students via knowledge representation networks (KRNets) with learnable parameters. Specifically, the texture-aware dynamic kernels are generated from local information to decompose the distillation problem into texture-wise supervision for further promoting the recovery quality of high-frequency details. In addition, the KRNets are optimized in a meta-learning manner to ensure the knowledge transferring and the student learning are beneficial to improving the reconstructed quality of the student. Experiments conducted on various single image super-resolution datasets demonstrate that our proposed method outperforms existing defined knowledge representation-related distillation methods and can help super-resolution algorithms achieve better reconstruction quality without introducing any extra inference complexity.     

Design and development of dynamic Agri-ontology for IoT interoperability

In recent years, the usage and applications of Internet of Things (IoT) have increased exponentially. IoT connects multiple heterogeneous devices like sensors, micro controllers, actuators, smart devices like mobiles, watches, etc. IoT contributes the data produced in the context of data collection, including the domains like military, agriculture, healthcare, etc. The diversity of possible applications at the intersection of the IoT and the web semantics has prompted many research teams to work at the interface between these two disciplines. This makes it possible to collect data and control various objects in transparent way. The challenge lies in the use of this data. Ontologies address this challenge to meet specific data needs in the IoT field. This paper presents the implementation of a dynamic agriculture ontology-building tool that parses the ontology files to extract full data and update it based on the user needs. The technology is used to create the angular library for parsing the OWL files. The proposed ontology framework would accept user-defined ontologies and provide an interface for an online updating of the owl files to ensure the interoperability in the agriculture IoT.     

A Dual-Mode Electrochromic Platform Integrating Zinc Anode-Based and Rocking-Chair Electrochromic Devices

The complementary electrochromic device, where the optical transmittance changes upon the flow of cations back and forth between anodic and cathodic electrodes, operates in a rocking-chair fashion if it can inherently self-discharge. Herein, the first demonstration of a dual-mode electrochromic platform having self-coloring and self-bleaching characteristics is reported, which is realized by sandwiching zinc metal within a newly-designed Prussian blue (PB)-WO₃ rocking-chair type electrochromic device. It is demonstrated that the redox potential differences between the zinc metal and the WO₃/PB electrodes endow the self-color-switching of these electrodes. By employing a hybrid electrolyte of Zn²⁺/K⁺, it is further shown that the colored PB-WO₃ rocking-chair device is capable of spontaneously bleaching when the anodic and cathodic electrodes are coupled. This dual-mode light-control strategy enables the electrochromic devices to exhibit four distinct optical states with the highest optical contrast of 72.6% and fast switching times (<5 s for the bleaching/coloration processes). Furthermore, the built-in voltage of the dual-mode electrochromic devices not only promotes energy efficiency, but also augments the bistability of the devices. It is envisioned that the broad implication of the present platform is in the development of self-powered smart windows, colorful displays, optoelectronic switches, and optical sensors.     

Disaster Monitoring of Satellite Image Processing Using Progressive Image Classification

The analysis of remote sensing image areas is needed for climate detection and management, especially for monitoring flood disasters in critical environments and applications. Satellites are mostly used to detect disasters on Earth, and they have advantages in capturing Earth images. Using the control technique, Earth images can be used to obtain detailed terrain information. Since the acquisition of satellite and aerial imagery, this system has been able to detect floods, and with increasing convenience, flood detection has become more desirable in the last few years. In this paper, a Big Data Set-based Progressive Image Classification Algorithm (PICA) system is introduced to implement an image processing technique, detect disasters, and determine results with the help of the PICA, which allows disaster analysis to be extracted more effectively. The PICA is essential to overcoming strong shadows, for proper access to disaster characteristics to false positives by operators, and to false predictions that affect the impact of the disaster. The PICA creates tailoring and adjustments obtained from satellite images before training and post-disaster aerial image data patches. Two types of proposed PICA systems detect disasters faster and more accurately (95.6%).     

Stealthy switching attacks on sensors against state estimation in cyber-physical systems

This article focuses on designing sensor attacks to deteriorate the state estimation in cyber-physical systems. The scenario that the malicious attack signals can be injected into different but limited number of sensor communication channels is considered. The state estimation error variations and innovation variations are adopted to measure attack performance and attack stealthiness, respectively. A switching attack strategy is proposed, under which the estimation error variations are driven to the predesigned target value and the norm of innovation variations remains at a small level. The switching attack design problem is formulated as a discrete switched optimal control problem which can be solved by dynamic programming, while the computational burden is heavy. To overcome this difficulty, by using pruning technique to remove the redundant matrices generated in dynamic programming, the quadratic optimization problem becomes numerically tractable. In this way, the suboptimal attack signal sequence and switching sequence can be acquired. Finally, a simulation example is provided to illustrate the effectiveness of the proposed attack strategy.     

Adaptive fuzzy leader-follower consensus control using sliding mode mechanism for a class of high-order unknown nonlinear dynamic multi-agent systems

In this article, a distributed leader-follower consensus approach is developed for a class of high-order unknown nonlinear dynamic multi-agent systems (MASs). Because every agent of the MAS contains multiple state variables, the existing consensus methods are not completely applicable for it. In order to find the qualified consensus protocol for this high-order MAS, sliding mode mechanism can be naturally considered for designing the consensus control because it can manage multiple state variables with the help of a constructed hyperplane. To this consensus control design, the sliding mode term is composed of all tracking error variables. Since the method does not require the switching control term around sliding surface, it can avoid the chattering phenomenon, which exits in most of the published sliding mode controls (SMCs). Furthermore, to handle the unknown nonlinear dynamic problem, the adaptive approximation strategy is implemented by employing fuzzy logic system (FLS). In the light of Lyapunov stability analysis, it is demonstrated that the proposed control approach can accomplish the consensus tasks. Finally, a numerical example is implemented to further show the desired results.     

Adaptive neural network dynamic event-triggered control for strong interconnected stochastic nonlinear systems with output constraint

This paper proposes a dynamic event-triggered mechanism based command filtered adaptive neural network (NN) tracking control scheme for strong interconnected stochastic nonlinear systems with time-varying output constraints. By designing a state observer, the unmeasured states of the systems can be estimated. The NNs are utilized to handle the unknown intermediate functions. In the controller design process, the asymmetric time-varying barrier Lyapunov functions are used to guarantee that the systems outputs do not violate the constraint regions. By integrating the command filter with variable separation technique, the controller design process is more simple, and the problem of algebraic-loop can be solved which caused by interconnected functions. According to the Lyapunov stability theory, it can be ensured that all signals of the systems are bounded in probability. Finally, the availability of the developed control scheme can be showed by the simulation example.     

振动抑制与能量俘获专刊序

工程系统中梁结构经常处于各种激励的作用下,因而梁结构在这种环境下不可避免地发生着各种各样的强迫振动.在梁结构发生振动的过程中,其自身会受到的温度、湿度、电磁场、裂纹等众多内外部因素的影响,而众多内外部因素就构成梁的多物理场耦合环境.在多场耦合环境下,Green函数法作为一种解析方法在研究梁的多场耦合振动问题方面具有优势,有利于讨论力、电、热、裂纹等因素作用下梁的振动特性和多场耦合特性.Green函数法相比于模态叠加法,优点在于能够得到完整且精度较高的解析解,具有收敛性好,运算快的特点.本文主要阐述Green函数在梁的强迫振动、热力耦合振动、力电耦合振动、裂纹梁振动等研究问题上取得了大量的理论和工程研究成果.本文以裂纹为内因,热、力、电为外因进行分类,阐述了在内外因影响下梁的强迫振动问题Green函数解的研究现状,从而让读者进一步系统性的了解Green函数法在振动领域中的广泛应用,以及了解该方法本身的特色和优势奠定基础.     

Control allocation technology based on fault diagnosis for the unmanned aerial vehicle system subject to physical constraints and fault reconfiguration mismatch

The problem of the system robustness subject to physical constraints and mismatched fault reconstruction is discussed in this paper. In order to facilitate the design, a four-rotor unmanned aerial vehicle (UAV) system model was selected for research. First, the control allocation model of the nonlinear UAV system with disturbances is shown in the paper. Secondly, a weighted pseudo-inverse method based on adaptive weights is proposed, which reduces the impact of physical constraints on the system. After that, a dynamic weight control allocation method based on the fault efficiency matrix is designed. The weight matrix can dynamically adjust the control distribution law according to the fault estimation value provided by the observer. Then, a dynamic adaptive control allocation method for faults and physical constraints is carried out by combining adaptive weights and dynamic weights. Finally, a simulation example is presented to further illustrate the effectiveness of the algorithm proposed in this paper.     

一种水下足式机器人的设计方案研究

海洋蕴含着丰富的自然资源,对海洋资源的合理开发与运用可以极大的缓解能源短缺问题,但由于海洋的复杂环境所带来的资源开发难度也不容忽略;针对这系列问题,对一种新型水下航行器进行研究,开展了水下足式机器人的总体设计分析,并通过三维建模对机器人进行结构设计;利用ANSYS对机器人关键运动部件进行应力分析,对机器人的设计进行完善;利用FLUENT软件进行流体模拟仿真,计算得到机器人腿部在水下运动时的受力特性;使用Adams软件对机器人的运动进行动力学仿真,计算机器人腿部扭矩输出特性;结合仿真结果对机器人的硬件系统进行选型,完成水下足式机器人的总体设计。