Control Theory and Technology

[Series 6 | ADRC – Core Theory & Design | #4] #ADRC #AircraftControl #FractionalOrder #ControlTheory ✈️🧠 How can a complex tailless aircraft be effectively decoupled and controlled? This work leverages a fractional-order error Extended State Observer (ESO) to achieve effective decoupling control in highly coupled dynamics. Title: Decoupling control for tailless aircraft based fractional-order error extended state observer Authors: Yunlong Hu, Mingfei Zhao, Jia Song, Wenling Li & Yang Liu Full text: rdcu.be/eESSm

[Series 6 | ADRC – Core Theory & Design | #3] #ADRC #Noise #Filters #ControlTheory 🌬️📉 Tired of sensor noise degrading ADRC performance? This work introduces a strategically placed low-pass filter with a clear tuning method, effectively reducing noise sensitivity—validated on a wind turbine system. 🤗😀 Title: Sensing-noise reduction in active disturbance rejection controllers: a permanent magnet synchronous generator-based wind turbine example Authors: Mario Andrés Aguilar-Orduña, Brian Camilo Gómez-León, Hebertt Sira-Ramírez, Rubén Alejandro Garrido-Moctezuma Full Text:rdcu.be/eG81P

[Series 6 | ADRC – Core Theory & Design | #2] #ADRC #Performance #EngineeringValidation #ControlTheory 🧲⚙️ How can disturbance compensation be further improved in linear ADRC? This work introduces a dual-loop compensation structure, inspired by an ideal integral chain model, to enhance ADRC performance—validated on a maglev platform. Title: LADRC method referring to the integral chain model-design of dual-loop disturbance compensation and engineering verification Authors: Yao Qin, Hailin Hu & Jie Yang More at: rdcu.be/eDjp7

[Series 6 | ADRC – Core Theory & Design | #1] 🌪️⚙️ Can ADRC handle a broader class of uncertain systems? This work develops a robust MP-ADRC strategy for minimum phase systems, requiring only the sign of the control gain. 🧐🧐🧐 Title: A robust MP-ADRC-based strategy for uncertain minimum phase systems Authors: Josiel A. Gouvêa, Alessandro R. L. Zachi, Lúcio M. Fernandes & Tiago Roux Oliveira More at: rdcu.be/eD32z #ADRC #RobustControl #MinimumPhase #ControlTheory

🔹 Now introducing Series 6: ADRC – Core Theory & Design 🌪️🛠️ How can we control systems effectively without relying on precise models? ⚙️ This #Thread dives into the core theory of Active Disturbance Rejection Control (ADRC)—from observer design and disturbance estimation to rigorous stability analysis. 💡 A powerful paradigm for handling uncertainty and rejecting disturbances in real-world systems. #ADRC #ControlTheory #RobustControl

[Series 5 | Machine Learning & Data-Driven Control | #9] #MedicalImaging #DeepLearning #HealthcareAI #ComputerVision 🧠🩻 How can medical images be segmented more accurately for diagnosis? This work proposes a multi-modality hierarchical fusion network, achieving 85.05% accuracy in segmenting fractured lumbar vertebrae from MRI scans. Title: Multi-modality hierarchical fusion network for lumbar spine segmentation with magnetic resonance images Authors: Han Yan, Guangtao Zhang, Wei Cui & Zhuliang Yu Full Text: rdcu.be/dVrsC

[Series 5 | Machine Learning & Data-Driven Control | #8] #GameTheory #Verification #STP #MultiAgentSystems 🧠🎲 How can complex multi-agent strategic interactions be rigorously verified? This work develops a matrix-based method using the Semi-Tensor Product (STP) to verify extended finite multi-potential games. Title: STP-based verification of extended finite multi-potential games Authors: Zhipeng Zhang, Haotian Peng, Jianbo Song & Zengqiang Chen Full Text: rdcu.be/epntJ

[Series 5 | Machine Learning & Data-Driven Control | #7] #ComputerVision #DeepLearning #Transformer #VisualTracking 🎯👁️ How can we reliably track fast-moving objects in cluttered environments? This work introduces a Convolution–Transformer Siamese network, combining CNN efficiency with Transformer global attention for robust, real-time visual tracking. 🧐🧐🧐 Title: Effective convolution mixed Transformer Siamese network for robust visual tracking Authors: Lin Chen, Yungang Liu & Yuan Wang Full Text: rdcu.be/eh4Ge

[Series 5 | Machine Learning & Data-Driven Control | #6] #AutonomousVehicles #ReinforcementLearning #EnergyEfficiency #DataDrivenControl 🚗⚡ How can autonomous driving be made more energy-efficient in urban environments? This work leverages Proximal Policy Optimization (PPO) to optimize driving behaviors at traffic signals, on-ramps, and in dense traffic, achieving improved energy efficiency. Title: Design of energy-saving driving strategy based on proximal policy optimization considering urban transport information Authors: Qifang Liu, Dazhen Sun, Haowen Chen, Dongzi Li & Ping Wang Full Text: rdcu.be/dZQav

[Series 5 | Machine Learning & Data-Driven Control | #5] #NeuralNetworks #ControlSystems #DeepLearning #DataDrivenControl 🧠⚖️ Can neural networks outperform traditional model-based controllers? This comparative study shows that feedforward and recurrent neural networks can surpass original analytical controller designs in overall control performance. 📝📝 Title: Application of feedforward and recurrent neural networks for model-based control systems Authors: Marek Krok, Wojciech P. Hunek, Szymon Mielczarek, Filip Buchwald, Adam Kolender Full text: https://‍rdcu.be/dZNpU

[Series 5 | Machine Learning & Data-Driven Control | #4] #NeuralNetworks #OptimalControl #ReinforcementLearning #DataDrivenControl 🧠⚙️ How can we build a self-learning controller for unknown multi-input systems? This work develops a data-based adaptive neural dynamic programming approach with integral reinforcement, avoiding the need to solve complex HJB equations. Title: Data-based neural controls for an unknown continuous-time multi-input system with integral reinforcement Authors: Yongfeng Lv, Jun Zhao, Wan Zhang, Huimin Chang Full Text: rdcu.be/d5hr3

[Series 5 | Machine Learning & Data-Driven Control | #3] #DataDrivenControl #LQR #OptimalControl #MachineLearning 📉🎯 Can optimal controllers be designed with far less experimental data? This work proposes a data-driven LQR method that reduces the required data amount by half, while maintaining strong control performance. 🧐🧐🧐 Title: Reduction of data amount in data-driven design of linear quadratic regulators Authors: Shinsaku Izumi, Xin Xin More at: rdcu.be/dUoscs

[Series 5 | Machine Learning & Data-Driven Control | #2] #SparseModeling #Optimization #SystemID #MachineLearning 🧩📉 Need a simpler and more interpretable model for complex dynamics? This work develops variable projection algorithms with sparse constraints to efficiently identify key parameters in separable nonlinear models. Title: Variable projection algorithms with sparse constraint for separable nonlinear models Authors: Hui-Lang Xu, Guang-Yong Chen, Si-Qing Cheng, Min Gan, Jing Chen Full Text: rdcu.be/dxg27

[Series 5 | Machine Learning & Data-Driven Control | #1] 🧠📈 Can machines learn hidden patterns from just a single sensor stream? Yes. This work uses observer-based deterministic learning for dynamic pattern recognition from univariate time series, with application to compressor stall prediction. 🧐Title: Dynamical pattern recognition for univariate time series and its application to an axial compressor 👥Authors: Jingtao Hu, Weiming Wu, Zejian Zhu and Cong Wang 👀More at: rdcu.be/dxTE9 #PatternRecognition #TimeSeries #MachineLearning #DataDrivenControl

🔹 Now introducing Series 5: Machine Learning & Data-Driven Control 🤖📊 What happens when control theory meets modern machine learning? ⚙️ This #Thread highlights how data-driven methods—from pattern recognition and predictive modeling to deep reinforcement learning—are tackling complex control problems and enabling smarter autonomous systems. #MachineLearning #DataDrivenControl #AI #ControlTheory

[Series 4 | System Identification & Estimation | #17] #Sensors #StochasticModeling #IMU #Estimation 📡📈 What is the best stochastic model for your IMU’s random errors? This work proposes an automatic model selection algorithm based on the Generalized Method of Wavelet Moments (GMWM) to identify the optimal stochastic process. ⚙️ Title: Automatic modeling algorithm of stochastic error for inertial sensors Authors: Luodi Zhao, Long Zhao Full Text: doi.org/10.1007/s11768…

[Series 4 | System Identification & Estimation | #16] #IntervalEstimation #Observer #TimeDelay #StateEstimation 📏🔍 How can states and noise be reliably estimated under disturbances and delays? This work develops an observer-based interval estimation method for time-delay discrete linear systems, providing guaranteed bounds on states and noise. ⚙️ Title: Observer design and interval estimation of time-delay discrete-time linear systems with external disturbance and measurement noise Authors: Junqi Yang, Jianhao Xie, Feiyang Liu Full Text: doi.org/10.1007/s11768…

[Series 4 | System Identification & Estimation | #15] #Benchmark #Modeling #Validation #SystemIdentification ⚙️🎯 How can a classic benchmark system be modeled from first principles and validated experimentally? This work presents a complete nonlinear model of the ball-and-plate system, with validation through correlation function analysis. 📊 Title: Mathematical modelling of ball and plate system with experimental and correlation function based model validation Authors: Dil Kumar T R, Mija S J* Full Text: rdcu.be/dHB8h

[Series 4 | System Identification & Estimation | #14] #Biomedical #Modeling #Physiology #SystemIdentification 🫁📈 How can we model the body’s oxygen response during changing exercise conditions? This work applies non-parametric kernel methods to capture the dynamics of oxygen uptake during switching stair exercise. 🧗‍♀️🩺🔍 Title: Oxygen uptake response to switching stairs exercise by non-parametric modeling Authors: Hairong Yu, Chenyu Zhang, Hamzah M Alqudah, Steven Su Full Text: rdcu.be/dHB5S

[Series 4 | System Identification & Estimation | #13] #MarineRobotics #OnlineLearning #SystemID #Estimation 🚤🧠 How can a vessel learn its own dynamics in real time? This work proposes an interactive system identification method with ESO-based disturbance estimation, enabling an unmanned surface vehicle to learn its motion model online. Title: Online interactive identification method based on ESO disturbance estimation for motion model of double propeller propulsion unmanned surface vehicle Authors: Yong Xiong, Xianfei Wang, Siwen Zhou Full Text: rdcu.be/dHB56