Slip detection plays a vital role in robotic manipulation and it has long been challenging problem the community. In this paper, we propose new method based on deep neural network (DNN) to detect slip. The training data is acquired by GelSight tactile sensor camera mounted gripper when use robot arm grasp lift 94 daily objects with different grasping forces positions. DNN trained classify whether slip occurred or not. To evaluate performance of DNN, test 10 unseen 152 grasps. A accuracy as...

This article presents a modified modular multilevel converter (MMC) equipped with the middle-cells and control algorithm for medium voltage applications. It is intended to reduce submodule (SM) capacitor ripples by third-order harmonic injection (THVI) second-order current (SHCI) without common-mode injected on ac-side. On one hand, using proposed algorithm, independent power degrees of freedom can be produced eliminate dominant fundamental part in arm powers, which dramatically SM ripples....

In response to the urgent demands of Internet Things for precise indoor target positioning and information interaction, this paper proposes a multi-band analog radio-over-fiber mobile fronthaul system. The objective is obtain target's location in environments while integrating remote beamforming capabilities achieve wireless access targets. Vector signals centered at 3, 4, 5, 6 GHz 30 are generated centrally distributed unit (DU) fiber-distributed active antenna (AAU) Target achieved by...

Traditional objective image quality evaluation measures, such as the MSE or PSNR, only represent total difference between original images and reconstructed images. However in some case, there are few large error pixels many small an image, they have not a consistent result with subjective measure. To cope this drawback, we propose new measure that is called geometry weighted separating block peak signal to noise ratio (GWSB/spl I.bar/PSNR). It corresponds human visual observations very well....

In this paper, a novel second-order active disturbance rejection control (2-ADRC) algorithm is proposed to optimize the of air supply subsystem for Proton Exchange Membrane Fuel Cell (PEMFC). To improve optimal effect PEMFC, modeling theory considering dynamic characteristics PEMFC system first studied, and Simulink model established verified. Then, oxygen excess ratio (OER) parameters under different load currents are obtained, OER also used as reference designed algorithms. addition,...

Based on the immune response and network theory, main mechanisms of interaction among antibody molecules, an artificial agent model system for multi-autonomous robots are put forward.By simulating experiments physical experiments, it indicates that presented cooperation approach can achieve a self-organizing is more robust flexible in dynamic environment.

Lithium-ion batteries are the core element in electric-vehicle systems. Because they provide a power source for entire electric vehicle, their state of charge (SOC) is crucial to ensure proper operation vehicle. Thus, more precise estimate SOC should significantly improve safety and utility these batteries. Toward this end, we propose herein novel nonlinear autoregressive with exogenous input (NARX) full-parallel (dual-close loop) structure NARX (FPNARX) networks overcome disadvantage...

ADVERTISEMENT RETURN TO ISSUEPREVViewpointNEXTPollution Control on River Nile in the White State Sugar Hub of SudanMohmmed Alnail†, Martin Kabenge*‡, Jianhua Li‡, Miao Zhu‡, Ali Abubaker§, and Aisha Magzoub∥View Author Information† Ministry Agriculture Forests, Government Sudan, P.O. Box 8178, Gadaref, 12217, Sudan‡ UNEP-Tongji Institute Environment for Sustainable Development, Key Laboratory Pollution Resource Reuse Study, Tongji University, Siping Rd 1239, Shanghai, 200092, China§...

Microelectromechanical systems (MEMS) suspended inductors have excellent radio frequency (RF) performance and they are compatible with integrated circuit (IC). They will be shocked during manufacturing, transportation, operation; in some applications, the shock amplitude can as high tens of thousands gravitational acceleration (g, 9.8 m/s2). High-g lead to inductor deformation which affects its or even failure structure. However, few studies been carried out on under high-g shock. In this...

Structural features of circulant matrices are utilized to study stabilities systems in this paper. They that a necessary and sufficient condition for asymptotic stability is given; when contain uncertain parameters, robust stable region determined. Besides, we present switched system model nonlinear model. For the former, explore under arbitrary switching laws with common quadratic Lyapunov function constructed, latter, find out locally asymptotically at origin. Finally, simulation examples...

In chip multiprocessor systems (CMPs), Network on Chip (NoC) has been widely used due to its advantages of favorable reusability, high reliability, and low power consumption. Recently, using NoC platforms accelerate deep neural networks (DNNs) become a new trend. This design can enable the intermediate computation results DNNs be transmitted within chip, reducing number accesses off-chip memory. However, large amount one-to-many traffic in DNN accelerator will occupy system bandwidth, which...

This paper describes a method of reducing the influence data error caused by non-ideal boundary shape and electrode placement on reconstructed images. A calibration modified Newton-Raphson reconstruction are combined in to ensure rapid convergence small error. The is verified simulated 2D software phantom practical measurements cylindrical saline tank.

Time series neural network (TSNNT) performs effectively in battery state of charge (SOC) estimation, but there are also some problems. Different data characteristics will affect the choice time delay value network, and prediction accuracy generality. The traditional RLS-based on-line open-circuit voltage (OCV) identification algorithms have shortcomings, which lead to larger errors obtained voltage, resulting SOC estimation. To solve these two problems, a novel estimation method based on is...

Thevenin equivalent circuit, providing a simple and feasible way to get the maximum transmission power load, is widely used in traditional ac systems for voltage instability identification. This paper makes some further study on this methodology with consideration of high direct current (HVDC) links. At first, an HVDC circuit founded quasi steady-state (QSS) model proposed, which can be extensively analysis long-term stability analysis. On basis, algorithms AC/DC are discussed. In order...

The mechanical properties of sandwich structures have drawn much attention from research and industrial application, while in the form L-joint been rarely addressed. In this paper, were experimentally tested under bending load to investigate performance during structural deformation damage evolution. corresponding testing process was numerically simulated using finite element (FE) modeling. Nomex® honeycomb selected as core material structure, with outer facesheets made thermosetting Carbon...

Microchip capillary electrophoresis (CE) has been recognized as a powerful tool for biochemical analyses due to its micro size, fast separation and low sample requirement. According the characters of CE signal, Db12 wavelet with level decomposition 5 thresholding method sqtwolog-hard were conformed optimum denoising strategy after computations. The result shows that, using this strategy, simulated noise could be removed adequately. Compared Fourier transform method, discrete wavelets (DWT)...