Article
Automation & Control Systems
Xiumei Han, Xudong Zhao, Tao Sun, Yuhu Wu, Ning Xu, Guangdeng Zong
Summary: This article examines the problem of event-triggered optimal control for discrete-time switched nonlinear systems with constrained control input, proposing a novel method called event-triggered heuristic dynamic programming (ETHDP) to derive optimal control policies effectively. The use of two neural networks helps decrease calculation and transmission load when the event-triggered condition is violated. The convergence of ETHDP is proven and the proposed method's effectiveness is verified through an example.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Automation & Control Systems
Chaoxu Mu, Ke Wang, Tie Qiu
Summary: This article introduces a novel event-triggering strategy integral reinforcement learning algorithm to reduce samples and transmission while ensuring learning performance. The core of the algorithm is policy iteration technique implemented by two neural networks.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Automation & Control Systems
Ping Wang, Zhen Wang, Xia Huang, Hao Shen
Summary: This paper investigates the asynchronous control of delayed switched neural networks via a dynamic event-triggering mechanism (DETM) and a merging signal scheme. The proposed improved DETM adds an exponential decaying term into the triggering condition, which has a larger lower bound compared to the static event-triggering mechanism (SETM) and the existing DETM. A merging signal is constructed by introducing a time-varying delay to combine the system switching signal with the controller switching signal, which allows for unified treatment of synchronous and asynchronous switching. Sufficient conditions for the asymptotical stability of the resulting closed-loop system are derived and a design algorithm for the feedback controller is provided.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Automation & Control Systems
Kun Jiang, Xuxi Zhang
Summary: This article investigates the event-triggered adaptive neural networks tracking control problem for nonlinear systems with prescribed performance and actuator fault. The study takes into account bias faults and the loss of effectiveness in actuators, while the effectiveness factor remains unknown. Neural networks are utilized to model the unknown terms of the systems during controller design using the backstepping technology framework, and an error transformation is employed to confine the tracking error within a predefined boundary. An adaptive neural networks event-triggered control strategy is developed to economize communication resources, ensuring that all closed-loop signals remain bounded and the tracking error asymptotically converges to zero. Two simulation examples are presented to validate the effectiveness of the proposed control strategy.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Multidisciplinary Sciences
Tao Yan, Rui Yang, Ziyang Zheng, Xing Lin, Hongkai Xiong, Qionghai Dai
Summary: Photonic neural networks use photons instead of electrons to perform brain-like computations, leading to significantly improved computing performance. However, current architectures are limited to handling data with regular structures and cannot generalize to graph-structured data beyond Euclidean space. In this study, a diffractive graph neural network (DGNN) is proposed to address this limitation by utilizing diffractive photonic computing units (DPUs) and on-chip optical devices. DGNN achieves complex feature representation by capturing dependencies among node neighborhoods during light-speed optical message passing over graph structures. It demonstrates superior performance in node and graph-level classification tasks with benchmark databases, providing a new direction for high-efficiency processing of large-scale graph data structures using deep learning.
Article
Biochemical Research Methods
Oezlem Muslu, Charles Tapley Hoyt, Mauricio Lacerda, Martin Hofmann-Apitius, Holger Froehlich
Summary: The study proposes a novel approach, GuiltyTargets, for prioritization of putative targets using attributed network representation learning and positive-unlabeled learning. The evaluation on multiple disease datasets demonstrates its superiority over previous methods and its potential for target repositioning across related diseases.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Automation & Control Systems
Qiliang Luo, Shan Xue, Derong Liu
Summary: In this article, a novel decentralised event-triggered control scheme is developed to stabilise interconnected systems. By restructuring interconnected systems in a distributed way and using adaptive critic designs with event-triggering mechanism, the robust control problem is transformed into an optimal control problem. The Lyapunov stability proof ensures the asymptotic stability of the whole system and the critic neural networks' ultimate boundedness.
INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE
(2022)
Article
Automation & Control Systems
Lijun Long, Fenglan Wang
Summary: This article presents a dynamic event-triggered adaptive neural network control approach for switched nonlinear systems. By utilizing switched command filter and common Lyapunov function method, the issues of asynchronous switching and discontinuous measurement error are addressed. Numerical examples demonstrate the effectiveness of the proposed approach.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Xu Huang, Bowen Zhang, Shanshan Feng, Yunming Ye, Xutao Li
Summary: In this paper, an interpretable local flow attention (LFA) mechanism is proposed for traffic flow prediction (TFP), which has the advantages of flow-awareness, interpretability, and efficiency. Based on LFA, a novel spatiotemporal cell called LFA-ConvLSTM is developed to capture the complex dynamics in traffic data. Experimental results demonstrate that our method outperforms previous approaches in prediction performance and is also faster by 32% than global self-attention ConvLSTM.
Article
Computer Science, Artificial Intelligence
Alejandro Moran, Vincent Canals, Fabio Galan-Prado, Christian F. Frasser, Dhinakar Radhakrishnan, Saeid Safavi, Josep L. Rossello
Summary: Edge artificial intelligence is a growing research field, and reservoir computing has attracted attention as a feasible alternative for edge intelligence. This study proposes a simple hardware-optimized circuit design for low-power edge intelligence applications and demonstrates its implementation in FPGA for low-power audio event detection. The results show significant accuracy and ultra-low energy consumption for the proposed approach.
COGNITIVE COMPUTATION
(2023)
Article
Computer Science, Information Systems
Haozhe Chen, Hang Zhou, Jie Zhang, Dongdong Chen, Weiming Zhang, Kejiang Chen, Gang Hua, Nenghai Yu
Summary: In recent years, various methods for protecting model intellectual property (IP) have been proposed, but the problem of quickly detecting copied models among a large number of models on the Internet has not received enough attention. This article introduces a novel model copy detection mechanism called perceptual hashing for convolutional neural networks (CNNs), which can efficiently retrieve similar versions of a query model by comparing hash codes. The experiment demonstrates the superior copy detection performance of the proposed perceptual hashing scheme on a model library of 3,565 models.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xingyue Liu, Kaibo Shi, Huaicheng Yan, Jun Cheng, Shiping Wen
Summary: This paper investigates a novel integral-based event-triggering switched control scheme for the LFC power system with an H∞ norm bound. To reduce redundant signals transmission, a novel integral-based event-triggering mechanism (IETM) is constructed. The IETM based switching control scheme is established, considering deception attacks. The less conservative H∞ asymptotic stability criterion is derived through delay-dependent Lyapunov-Krasovskii functionals with integral terms and the controller design method based on the IETM switching control rule is obtained. The effectiveness and superiority of the designed IETM are validated based on a wind-connected power system example.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Chemistry, Analytical
Shouyan Chen, Mingyan Zhang, Xiaofen Yang, Zhijia Zhao, Tao Zou, Xinqi Sun
Summary: This paper discusses the applicable rules of Global-Attention and Self-Attention in SER classification construction, and proposes a new classifier model with an accuracy of 85.427% on the EMO-DB dataset.
Article
Computer Science, Artificial Intelligence
Leander Weber, Sebastian Lapuschkin, Alexander Binder, Wojciech Samek
Summary: Explainable Artificial Intelligence (XAI) is a research field that aims to bring transparency to complex and opaque machine learning models. This paper provides an overview of techniques that practically apply XAI to improve ML models, categorizing and comparing their strengths and weaknesses. Theoretical perspectives and empirical experiments demonstrate how explanations can enhance properties such as model generalization and reasoning. The potential caveats and drawbacks of these methods are also discussed.
INFORMATION FUSION
(2023)
Article
Automation & Control Systems
Songlin Hu, Zihao Cheng, Dong Yue, Chunxia Dou, Yusheng Xue
Summary: This paper proposes a bandwidth allocation-based switched dynamic triggering control method, which effectively prevents the system instability caused by DoS attacks and time delay by introducing a primary-redundancy communication structure and a dynamic event-triggered mechanism.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Engineering, Electrical & Electronic
Xinjun Wang, Ben Niu, Huanqing Wang, Xudong Zhao, Wendi Chen
Summary: This article focuses on the adaptive bipartite consensus issue of nonlinear multi-agent systems in directed graphs from a new perspective. A new distributed control algorithm, named finite-time prescribed performance control, is designed by using a prescribed performance function and a novel first-order filter. This algorithm ensures that the bipartite consensus errors converge to a prescribed compact set within a finite time and allows the system to achieve the prescribed performance and fast finite-time convergence. Furthermore, neural networks are introduced to handle the continuous unknown nonlinearity and the effect of non-strict feedback structure in the system, while a dynamic surface control mechanism with a novel first-order filter is used to overcome the complexity explosion problem in controller design. Simulation experiments on forced damped pendulums are conducted to demonstrate the feasibility of the theoretical work.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2023)
Article
Computer Science, Artificial Intelligence
Huanqing Wang, Jiawei Ma, Xudong Zhao, Ben Niu, Ming Chen, Wei Wang
Summary: This article considers the adaptive fuzzy fixed-time fault-tolerant tracking control problem for high-order nonlinear systems (HONSs) with sensor and actuator faults. Fuzzy logic systems are used to approximate the unknown nonlinear functions of the HONS. Based on backstepping technology and fixed-time theory, an adaptive fuzzy fixed-time fault-tolerant controller is developed to ensure bounded signals of the closed-loop HONS. A numerical example is presented to demonstrate the rationality of the proposed method.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Automation & Control Systems
Li-Min Han, Wei Su, Ben Niu, Xiao-Mei Wang, Xiao-Mei Liu
Summary: This paper proposes an adaptive compensation control algorithm to solve the actuator failures problem of nonlinear stochastic multi-agent systems under directed communication topology. Fuzzy logic systems are used to deal with the unknown nonlinearities, and a threshold-based event-triggered mechanism is considered to reduce communication burden. The dynamic surface control technique is also used to solve the issue of complexity explosion. Simulation results demonstrate the validity of the proposed design scheme.
IET CONTROL THEORY AND APPLICATIONS
(2023)
Article
Automation & Control Systems
Hao Jiang, Xiaomei Wang, Ben Niu, Huanqing Wang, Xinyu Liu
Summary: This article focuses on the event-triggered adaptive tracking containment control problem for a class of nonlinear multi-agent systems. To tackle the difficulties caused by unknown nonlinearities and unmodeled dynamics, Gaussian function properties and novel dynamics signals are used. A relative threshold-based event-triggered mechanism is also adopted to reduce system communication burden. The proposed protocol ensures convergence of followers' outputs to the convex hull spanned by the leaders' outputs, uniformly ultimate boundedness of all signals in the closed-loop system, and effective avoidance of Zeno behavior. Simulation results are provided to validate the effectiveness of the proposed containment control protocol.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Madini O. Alassafi, Muhammad Sohail Ibrahim, Imran Naseem, Rayed AlGhamdi, Reem Alotaibi, Faris A. Kateb, Hadi Mohsen Oqaibi, Abdulrahman A. Alshdadi, Syed Adnan Yusuf
Summary: The vulnerability of conventional face recognition systems to face presentation or face spoofing attacks has attracted attention. Deep learning-based face presentation attack detection (PAD) methods have gained popularity. This research proposes a supervised contrastive learning approach to tackle the face anti-spoofing problem.
APPLIED INTELLIGENCE
(2023)
Article
Mathematics, Interdisciplinary Applications
M. I. N. G. Z. H. U. Tang, C. A. I. H. U. A. Meng, L. A. N. G. LI, H. U. A. W. E. Wu, Y. A. N. G. Wang, J. U. N. B. I. N. He, Y. U. J. I. E. Huang, Y. U. Yu, M. A. D. I. N. I. O. Alassafi, F. A. W. A. Z. E. Alsaadi, A. D. I. L. M. Ahmad, F. U. Q. I. A. N. G. Xiong
Summary: An improved Borderline-SMOTE oversampling method called TSDAS-SMOTE is proposed to address the class-imbalance issue in wind turbine pitch connecting bolt data. TSDAS-SMOTE, combined with XGBoost, is used to construct a fault detection model. Experimental results show that the proposed method outperforms six popular oversampling methods in terms of missed alarm rate (MAR) and false alarm rate (FAR), achieving effective fault detection for large wind turbine pitch connection bolts.
FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY
(2023)
Article
Mathematics, Interdisciplinary Applications
Xue Tian, Madini O. Alassafi, Fawaz E. Alsaadi
Summary: The cultivation of creativity is closely related to language learning, and the challenge faced by English teachers is how to design the creativity promotion mechanism of English teaching in the public environment. With the advent of the era of big data, English teachers can apply the latest research results to classroom teaching, using social media to help students learn and communicate in the language, cultivate their creativity in learning English, and improve the quality of teaching.
FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY
(2023)
Article
Mathematics, Interdisciplinary Applications
Mingzhu Tang, Jun Tang, Huawei Wu, Yang Wang, Yiyun Hu, Beiyuan Liu, Madini O. Alassafi, Fawaz E. Alsaadi, Adil M. Ahmad, Fuqiang Xiong
Summary: Abnormal detection of wind turbine converter is a crucial technology for ensuring the stable operation and safe power generation of wind turbines. To address the issue of limited abnormal data and low recognition rate, a sample enhancement method based on an improved conditional Wasserstein generative adversarial network is proposed. Experimental results demonstrate that this method achieves lower MAR and FAR in the anomaly detection of wind turbine converters, outperforming other oversampling data balance optimization methods such as SMOTE, RandomOverSampler, GAN, etc.
FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY
(2023)
Article
Automation & Control Systems
Peng Wang, Hehong Zhang, Hong Sang, Ben Niu
Summary: This work focuses on the control problem of switched positive systems and proposes a co-design of controllers and a switching strategy. A hysteresis switching strategy is devised by dividing the nonnegative state space into subdomains, resulting in reduced switching frequency and allowing unsolvable almost output tracking control problems. The establishment of almost output tracking criteria based on multiple linear copositive Lyapunov functions ensures asymptotic convergence of the tracking error and prescribed L-1-gain index. The findings are applied to the output tracking control of a boost converter circuit.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Automation & Control Systems
Zhongwen Cao, Ben Niu, Guangdeng Zong, Xudong Zhao, Adil M. M. Ahmad
Summary: This article investigates the active disturbance rejection-based distributed event-triggered bipartite consensus problem of nonaffine nonlinear multiagent systems with input saturation. An event-triggered mechanism is employed for each follower to reduce the update frequency of the control signal. The active disturbance rejection technology, a combination of the extended state observer and the tracking differentiator, is introduced to estimate uncertainties and address complexity issues in the control law design.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Automation & Control Systems
Fabin Cheng, Ben Niu, Ning Xu, Xudong Zhao, Adil M. Ahmad
Summary: This paper proposes a low-computation design scheme for fault detection and performance recovery based on deferred replacement actuators for a class of uncertain nonlinear systems. The proposed method does not require prior knowledge of fault models, nor does it require multiple actuators working in parallel to mitigate the impact of faults. It achieves performance recovery by designing fault detection and shifting functions, and establishes a computationally efficient design scheme.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Automation & Control Systems
Ben Niu, Bocheng Yan, Xudong Zhao, Baoyi Zhang, Tao Zhao, Xiaomei Liu
Summary: This paper investigates the event-triggered-based adaptive bipartite finite-time tracking control problem of nonlinear nonstrict-feedback coopetition multi-agent systems (MASs) with time-varying disturbances. The major design difficulties are solved by utilizing radial basis function neural networks and Gaussian functions. The proposed control approach successfully drives the tracking errors to the desired neighborhood of the origin in an almost fast finite time.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Automation & Control Systems
Haiyang Chen, Guangdeng Zong, Xiang Liu, Xudong Zhao, Ben Niu, Fangzheng Gao
Summary: This paper investigates the attack-compensated output control problem in Markov jump cyber-physical systems subject to mismatched modes. An adaptive probabilistic event-triggered mechanism is developed to enhance the control performance of the networked control system. A predictor-based compensator is constructed to mitigate the impact of attacks on the control performance. A mismatched output feedback controller is designed, and the stability analysis is performed. Simulations are conducted to validate the proposed results.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Computer Science, Information Systems
Ayesha, Muhammad Javed Iqbal, Iftikhar Ahmad, Madini O. Alassafi, Ahmed S. Alfakeeh, Ahmed Alhomoud
Summary: This research focuses on comprehensive methodology of tiny vehicle detection using Deep Neural Networks (DNN) and achieves better performance compared to other SOTA techniques on KITTI benchmark dataset.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Sabbir Ahmed, Mohammad Abu Yousuf, Muhammad Mostafa Monowar, Abdul Hamid, Madini O. Alassafi
Summary: Depression and anxiety are common mental illnesses that are often overlooked. The current research primarily focuses on one or two factors for detection purposes, failing to consider all possible factors. To address this, researchers have developed a multimodal diagnosis system and proposed an attention-based multimodal classifier that can effectively train different modal datasets. Experimental results have shown that this approach achieves high accuracy, although missing modalities in the model may result in uncertainty in predictions.