Browsing by Author "Zhang, Huiyan"

Now showing 1 - 5 of 5

A novel electronic chip detection method using deep neural networks
(2022) Zhang, Huiyan; Sun, Hao; Peng, Shi; Minchala Ávila, Luis Ismael
Electronic chip detection is widely used in electronic industries. However, most existing detection methods cannot handle chip images with multiple classes of chips or complex backgrounds, which are common in real applications. To address these problems, a novel chip detection method that combines attentional feature fusion (AFF) and cosine nonlocal attention (CNLA), is proposed, and it consists of three parts: a feature extraction module, a region proposal module, and a detection module. The feature extraction module combines an AFF-embedded CNLA module and a pyramid feature module to extract features from chip images. The detection module enhances feature maps with a region intermediate feature map by spatial attentional block, fuses multiple feature maps with a multiscale region of the fusion block of interest, and classifies and regresses objects in images with two branches of fully connected layers. Experimental results on a medium-scale dataset comprising 367 images show that our proposed method achieved mAP0.5 = 0.98745 and outperformed the benchmark method.
Adaptive neural security control for networked singular systems under deception attacks
(2022) Ao, Wengang; Minchala Ávila, Luis Ismael; Zhao, Ning; Zhang, Huiyan
This paper studies the issue of the adaptive neural security controller design for uncertain networked singular systems in the presence of deception attacks. Considering that the attack signal is unknown, the neural networks technique is exploited to approximate the attack signal, which eliminates the assumption that the attack signal has a known upper bound. By combining the state feedback with the estimated information of the attack, the impact of the attack is effectively compensated. Furthermore, a novel Lyapunov function, including the decomposed state vector and the weight matrix estimation error, is established to evaluate the bounded area of the system state. Finally, a numerical example substantiates the validity of the theoretical results
Design and evaluation of an energy management system applied to a lower limb robotic exoskeleton
(2021) Zhang, Huiyan
This paper presents the design, development and evaluation of an energy management system (EMS) for covering the energy demand from an autonomous lower limb exoskeleton (ALLEX) prototype. The ALLEX prototype is composed by four energy subsystems: actuators, sensing, communications, and control. The energy de-manded by ALLEX is estimated by considering metabolic requirements of neurological rehabilitation applied to the actuators subsystem, as well as average consumption of the sensing, communications, and control subsystems. The EMS proposed in this paper is composed by a lithium-ion battery bank, a battery management system (BMS), and prop-er instrumentation for measuring voltages, currents, and temperature from the battery pack. Experimental results show adequate coverage of the energy demand from ALLEX, both instantly and during continuous operation (1 hour approximately). Additionally, the efficiency of the EMS is assessed by testing the cells balancing and battery charg-ing/discharging processes, which showed equalized values of the energy cells as well as correct temperature operating values
Dissipative output feedback control for semi-Markovian jump systems under hybrid cyber-attacks
(2021) Zhang, Huiyan; Shi, Peng; Minchala Ávila, Luis Ismael; He, Xiuli
In this paper, the dissipativity-based dynamic output feedback controller (DOFC) design for Semi-Markovian jump systems under stochastic cyber-attacks is first proposed. It is assumed that the time-varying uncertainties obey Bernoulli-distribution and transition probability matrix is time-varying and partially accessed. By utilizing the dissipativity-based technique, sufficient conditions for the existence of the DOFC are obtained to ensure the exponential stability with a strict dissipative performance of the resulted system. Next, the proposed results are improved by fractionalizing the time-varying transition probability matrix and the corresponding DOFC gains are obtained by cone complementarity linearization algorithm. Simulations results are provided to demonstrate the effectiveness and theoretical value of the proposed dissipativity-based DOFC design method
Time-limited model reduction for semi-markovian jump systems based on generalized gramians
(2021) Lan, Pin; Minchala Ávila, Luis Ismael; Li, Huimin; Zhang, Huiyan
This paper is concerned with the approximation problem for semi-Markovian jump systems (SMJSs) based on generalized time-limited Gramians. For a given SMJS with partially known transition probabilities, the generalized time-limited Gramians are defined and then solved by using time-interval input/output energy functions. The proposed model reduction algorithm can solve the synchronization jump problem, and also achieve the required reduced accuracy within the expected time-interval. Specially, the reduced error of the proposed algorithm is lower compared with the balanced truncation method, which has more theoretical value in time-interval control synthesis. Finally, a numeral example including four subsystems is given to estimate the effectiveness of the proposed results. © 2021 ICIC International.