Article
Automation & Control Systems
Jiang Li, Xiaoping Wang, Zhigang Zeng
Summary: In real-world dialog systems, understanding user emotions and interacting anthropomorphically is crucial. Emotion Recognition in Conversation (ERC) is a key approach to achieve this goal and has gained increasing attention. This study proposes a model called DualRAN, which combines recurrent and attention mechanisms to model conversations. Experimental results show that DualRAN achieves competitive performance on multiple datasets.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Tran Cong Dao, Nguyen Thi Tam, Huynh Thi Thanh Binh
Summary: Wireless Sensor Networks (WSNs) face challenges related to limited energy resources, and network lifetime and energy consumption are critical considerations. This paper introduces a novel approach to extend network lifetime and reduce energy consumption in WSNs by optimizing network architecture selection. The proposed method addresses limitations of previous studies and consistently generates valid solutions by incorporating efficient encoding and tailor-made genetic operators. It also harnesses knowledge transfer in a multitask evolutionary algorithm to explore various network architectures and achieve state-of-the-art results in terms of solution quality, convergence rate, and running time.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Halar Memon, Eskil Gjerde, Alex Lynam, Amiya Chowdhury, Geert De Maere, Grazziela Figueredo, Tanvir Hussain
Summary: This study proposes the first use of the active learning framework in thermal spray to enhance the accuracy of in-flight particle characteristics prediction. By implementing Bayesian Optimization, the maximum uncertainty is reduced, significantly improving the prediction accuracy and informativeness of the existing database. The AL-driven optimization not only accurately predicts the particle characteristics but also finds expected improvements around desired in-flight characteristics.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Elena Petri, Romain Postoyan, Daniele Astolfi, Dragan Nesic, W. P. M. H. (Maurice) Heemels
Summary: This study investigates a scenario where a perturbed nonlinear system transmits its output measurements to a remote observer via a packet-based communication network. By designing both the observer and the local transmission policies, accurate state estimates can be obtained while only sporadically using the communication network.
Article
Automation & Control Systems
Jonas Krook, Robi Malik, Sahar Mohajerani, Martin Fabian
Summary: This paper proposes a method to synthesise controllers for cyber-physical systems subjected to disturbances, such that the controlled system satisfies specifications given as linear temporal logic formulas. The approach constructs a finite-state abstraction of the original system and synthesises a controller for the abstraction. It introduces the robust stutter bisimulation relation to account for disturbances and uncertainty, ensuring that related states have similar effects under the same controller. The paper demonstrates that the existence of a controller for the abstracted system implies the existence of a controller for the original system enforcing the linear temporal logic formula.
Article
Automation & Control Systems
Clement Chahbazian, Karim Dahia, Nicolas Merlinge, Benedicte Winter-Bonnet, Aurelien Blanc, Christian Musso
Summary: The paper derives a recursive formula of the Fisher information matrix on Lie groups and applies it to nonlinear Gaussian systems on Lie groups for testing. The proposed recursive CRLB is consistent with state-of-the-art filters and exhibits representative behavior in estimation errors. This paper provides a simple method to recursively compute the minimal variance of an estimator on matrix Lie groups, which is fundamental for implementing robust algorithms.
Article
Automation & Control Systems
Jicong Duan, Xibei Yang, Shang Gao, Hualong Yu
Summary: The study proposes a novel partition-based imbalanced multi-label learning algorithm, MLHC, which divides the original label space into disconnected subspaces using hierarchical clustering. It successfully tackles the class imbalance problem in multi-label data and outperforms other class imbalance multi-label learning algorithms.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Yagang Zhang, Xue Kong, Jingchao Wang, Siqi Wang, Zheng Zhao, Fei Wang
Summary: This paper introduces a comprehensive wind speed forecasting system, including a signal reconstruction system, a signal reconstruction prediction model, and an error analysis algorithm. Experimental results show that the system has good performance and accuracy in wind speed prediction.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Jongchan Baek, Changhyeon Lee, Young Sam Lee, Soo Jeon, Soohee Han
Summary: This work utilizes reinforcement learning to achieve real-time control of a non-simulated triple inverted pendulum, using a structure-aware virtual experience replay method to enhance learning efficiency, and demonstrates its effectiveness on an actual system.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Dexuan Zou, Mengdi Li, Haibin Ouyang
Summary: In this study, a photovoltaic thermal collector is integrated into a combined cooling, heating, and power system to reduce primary energy consumption, operation cost, and carbon dioxide emission. By applying a novel genetic algorithm and constraint handling approach, it is found that the CCHP scenarios with PV/T are more efficient and achieve the lowest energy consumption.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Automation & Control Systems
Philippe Schuchert, Vaibhav Gupta, Alireza Karimi
Summary: This paper presents the design of fixed-structure controllers for the As2 and Asw synthesis problem using frequency response data. The minimization of the norm of the transfer function between the exogenous inputs and performance outputs is approximated through a convex optimization problem involving Linear Matrix Inequalities (LMIs). A general controller parametrization is used for continuous and discrete-time controllers with matrix transfer function or state-space representation. Numerical results show that the proposed data-driven method achieves performance equivalent to model-based approaches when a parametric model is available.
Article
Automation & Control Systems
Te Zhang, Bo Zhu, Lei Zhang, Qingrui Zhang, Tianjiang Hu
Summary: This paper introduces a control technique called time-varying uncertainty and disturbance estimator (TV-UDE) which extends the classic UDE approach to handle more complicated issues. By combining TV-UDE with a nominal dynamic output-feedback controller, robust control for uncertain second-order attitude control systems without velocity measurements is achieved. Numerical simulations and physical experiments on a 2-DOF AERO attitude helicopter platform demonstrate the effectiveness of the proposed design in reducing steady-state errors and avoiding issues caused by high-gain estimation.
CONTROL ENGINEERING PRACTICE
(2024)
Article
Automation & Control Systems
Matteo Della Rossa, Thiago Alves Lima, Marc Jungers, Raphael M. Jungers
Summary: This paper presents new stabilizability conditions for switched linear systems with arbitrary and uncontrollable underlying switching signals. The study focuses on two specific settings: the robust case with completely unknown and unobservable active mode, and the mode-dependent case with controller depending on the current active switching mode. The technical developments are based on graph-theory tools and path-complete Lyapunov functions framework, enabling the design of robust and mode-dependent piecewise linear state-feedback controllers using directed and labeled graphs.
Article
Automation & Control Systems
Nitin K. Singh, Abhisek K. Behera
Summary: In this paper, a twisting observer is proposed for robustly estimating the states of a second-order uncertain system. The observer approximates the unknown sign term for the non-measurable state with a delayed output-based switching function, and achieves the desired steady-state accuracy by controlling the delay parameter. The application of the observer to output feedback stabilization is also discussed.
Article
Automation & Control Systems
Noureddine Toumi, Roland Malhame, Jerome Le Ny
Summary: This paper addresses large multi-agent dynamic discrete choice problems using a linear quadratic mean field games framework. The model incorporates the features where agents have to reach a predefined set of possible destinations within a fixed time frame and running costs can become negative to simulate crowd avoidance. An upper bound on the time horizon is derived to prevent agents from escaping to infinity in finite time. The existence of a Nash equilibrium for infinite population and its epsilon-Nash property for a large but finite population are established. Simulations are conducted to explore the model behavior in various scenarios.
Article
Automation & Control Systems
Xiaoyu Luo, Chengcheng Zhao, Chongrong Fang, Jianping He
Summary: This paper investigates the problem of false data injection attacks in multi-agent dynamical systems and proposes FDI attack set selection algorithms to maximize the convergence error by finding the optimal subset of compromised agents.
Article
Automation & Control Systems
Weijun Ma, Bo Yang, Yuanshi Zheng
Summary: This paper develops new practical stability criteria for impulsive stochastic functional differential systems with distributed-delay dependent impulses, and shows that under certain conditions, the practical exponential stability of the systems remains unchanged.
NONLINEAR ANALYSIS-HYBRID SYSTEMS
(2024)
Article
Automation & Control Systems
Taher Ebrahim, Sankaranarayanan Subramanian, Sebastian Engell
Summary: We propose a robust nonlinear model predictive control algorithm for dynamic systems with mixed degrees of freedom. This algorithm optimizes both continuous and discrete manipulated variables, enhancing closed-loop performance. Our approach relies on a computationally efficient relaxation and integrality restoration strategy and provides sufficient conditions to establish recursive feasibility and guarantee robust closed-loop stability. The effectiveness of the approach is demonstrated through two nonlinear simulation examples.
Article
Automation & Control Systems
Jin Gyu Lee, Rodolphe J. Sepulchre
Summary: This paper examines how weak synaptic coupling can achieve rapid synchronization in heterogeneous networks. The assumptions aim to capture the key mathematical properties that make this possible for biophysical networks. The combination of nodal excitability and synaptic coupling are shown to be essential to the phenomenon.
Review
Automation & Control Systems
Qing Qin, Yuanyuan Chen
Summary: This paper offers a comprehensive review of retinal vessel automatic segmentation research, including both traditional methods and deep learning methods. In particular, supervised learning methods are summarized and analyzed based on CNN, GAN, and UNet. The advantages and disadvantages of existing segmentation methods are also outlined.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)