Explore the vast range of titles available.

In cyber-physical systems governed by nonlinear partial differential equations (PDEs), real-time control is often limited by sparse sensor data and high-dimensional system dynamics. Deep reinforcement learning (DRL) has shown promise for controlling such systems, but training DRL agents directly on full-order simulations is computationally intensive. This paper presents a sensor-driven, non-intrusive reduced-order modeling (NIROM) framework called FAE-CAE-LSTM, which combines convolutional and fully connected autoencoders with a long short-term memory (LSTM) network. The model compresses high-dimensional states into a latent space and captures their temporal evolution. A DRL agent is trained entirely in this reduced space, interacting with the surrogate built from sensor-like spatiotemporal measurements, such as pressure and velocity fields. A CNN-MLP reward estimator provides data-driven feedback without requiring access to governing equations. The method is tested on benchmark systems including Burgers’ equation, the Kuramoto–Sivashinsky equation, and flow past a circular cylinder; accuracy is further validated on flow past a square cylinder. Experimental results show that the proposed approach achieves accurate reconstruction, robust control, and significant computational speedup over traditional simulation-based training. These findings confirm the effectiveness of the FAE-CAE-LSTM surrogate in enabling real-time, sensor-informed, scalable DRL-based control of nonlinear dynamical systems.

This article deals with experimental identification and control of laboratory helicopter model CE 150 manufactured by company Humusoft. Structure of the identified system was approximated by linear black-box models. Discrete Input/Output Auto-Regressive Moving Average model with eXternal input (ARMAX) and its state space equivalent were used. Parameters of the models were estimated by regression techniques using System Identification Toolbox for Matlab. Acquired models were validated using simulations, residual analysis and real-time control. Input/output data necessary for identification were obtained by measurements from laboratory model and were processed using Real-Time Toolbox for Matlab. Based on acquired mathematical models input/output and state space controllers were designed (input/output pole placement with integration, state space pole placement with integration and observer). Designed controllers were implemented in Matlab environment using the Real Time Toolbox and their performance was verified by real-time control of the helicopter model.

The focus of this research was on building a decision support system for a model that characterizes the conflict interaction of n-dimensional complex systems with non-trivial internal structures. The interpretation of the new model was focused on information warfare as the impact of rare events that quickly change certain perceptions of a large number of people. Consequently, the support for various ideas experiences stochastic jumps, a phenomenon observable through a non-classical Levy approximation scheme. The essence of our decision support system lies in its ability to navigate the complex dynamics of conflict interaction among multifaceted systems. Through the utilization of advanced modeling techniques, our aim is to illuminate the complicated interplay of factors influencing information warfare and its cascading effects on societal perceptions and behaviors. Key components of our decision support system encompass model development, simulation capabilities, data integration, and visualization tools. The significance of our work lies in its potential to inform policy formulation, conflict resolution strategies, and societal resilience in the face of information warfare. By providing decision-makers with actionable intelligence and foresight into emerging threats and opportunities, our decision support system serves as a valuable tool for navigating the complexities of modern conflict dynamics. In conclusion, developing a decision support system for modeling conflict interaction in complex systems represents an essential step toward enhancing our understanding of information warfare and its consequences. Through interdisciplinary collaboration and innovative modeling techniques, we aim to provide stakeholders with the insights and capabilities needed to navigate the developing landscape of conflict and ensure the stability and resilience of society.

We present a method to solve fractional optimal control problems, where the dynamic control system depends on integer order and Caputo fractional derivatives. Our approach consists in approximating the initial fractional order problem with a new one that involves integer order derivatives only. The latter problem is then discretized, by application of finite differences, and solved numerically. We illustrate the effectiveness of the procedure with an example.

This article presents a control scheme for navigation by waypoints given in the X-Y plane for an underactuated hovercraft, which has three Degrees of Freedom and two control inputs. The control scheme is made up of three feedforward Backstepping Controllers (BC); two act directly on the actuated coordinates, while the third corresponds to an auxiliary control that is responsible for manipulating the orientation angle (actuated coordinate) based on the error of the Y axis (underactuated coordinate), allowing the tracking of desired reference points in the X-Y plane. The proposed scheme is numerically validated for different cases that involve trajectory tracking through different waypoints. Additionally, the displacement from one reference point to another is carried out by tracking a Bezier polynomial trajectory, which guarantees a smooth transition from one waypoint to another; therefore, to controlling the time in which the vehicle reaches the reference value. Finally, a comparison of the proposed scheme with respect to the use of Feedforward PID Controllers (FFPC) is presented.

This paper presents a comprehensive tutorial on the identification process for a class of continuously dynamic systems expressed in state-space form using the model reference adaptive approach. The proposed algorithm does not require prior knowledge of the systems but needs all state variables to adapt the estimated parameters. Through simulations using m-script code, software-in-the-loop (SIL), and hardware-in-the-loop (HIL) simulations, the effectiveness of the proposed method in identifying the system model of a DC motor is evaluated. Simulation results demonstrate consistency across various platforms. Steady-state estimated models can be achieved using the proposed estimation algorithm with adaptation gains of diag ( [ 100 100 ] ) after 5 s. Furthermore, this paper demonstrates the implementation of the proposed method on both SIL and HIL platforms, using Python and MicroPython programming languages, respectively. This approach leverages the Numpy library for efficient matrix computations. It is evident that the proposed estimation algorithm is readily applicable in real-world scenarios.

We propose an interdisciplinary framework for time series classification, forecasting, and anomaly detection by combining concepts from Koopman operator theory, machine learning, and linear systems and control theory. At the core of this framework is nonlinear dynamic generative modeling of time series using the Koopman operator which is an infinite-dimensional but linear operator. Rather than working with the underlying nonlinear model, we propose two simpler linear representations or model forms based on Koopman spectral properties. We show that these model forms are invariants of the generative model and can be readily identified directly from data using techniques for computing Koopman spectral properties without requiring the explicit knowledge of the generative model. We also introduce different notions of distances on the space of such model forms which is essential for model comparison/clustering. We employ the space of Koopman model forms equipped with distance in conjunction with classical machine learning techniques to develop a framework for automatic feature generation for time series classification. The forecasting/anomaly detection framework is based on using Koopman model forms along with classical linear systems and control approaches. We demonstrate the proposed framework for human activity classification, and for time series forecasting/anomaly detection in power grid application.

The effective tuning of Proportional-Integral-Derivative (PID) controllers is crucial for industrial process control, but achieving optimal parameters for complex systems remains challenging. The recent Artificial Satellite Search Algorithm (ASSA) is strong in exploration but suffers from an imbalance between global and local search and a greedy selection strategy, leading to premature convergence. To overcome these limitations, this paper proposes an enhanced variant, MEASSA (Memory-based and Evolutionary-enhanced ASSA), which integrates a memory mechanism to preserve elite solutions, an evolutionary operator for guided population dynamics, and a stochastic local search for intensive refinement. Experimental evaluations on three dynamic systems are a DC motor, a three-tank liquid level system, and a fourth-order system which demonstrate MEASSA’s superior performance. The algorithm achieved the lowest Integral Absolute Error (IAE) values of 9.977, 9.0781, and 9.697, respectively, outperforming several benchmark metaheuristics. Time-domain and frequency-domain analyses further confirm its ability to minimize overshoot, improve settling time, and enhance system stability, validating MEASSA as a robust and accurate method for complex PID controller tuning.

In this study, a nonlinear dynamic model is developed to examine the stability and vibration behavior of a self-balancing electric Segway operating over irregular stochastic terrains. The Segway is treated as a three-degrees-of-freedom cart–inverted pendulum system, incorporating elastic and damping effects at the wheel–ground interface. Road irregularities are generated in accordance with international standard using high-order filtered noise, allowing for representation of surface classes from smooth to highly degraded. The governing equations, formulated via Lagrange’s method, are transformed into a Lorenz-like state-space form for nonlinear analysis. Numerical simulations employ the fourth-order Runge–Kutta scheme to compute translational and angular responses under varying speeds and terrain conditions. Frequency-domain analysis using Fast Fourier Transform (FFT) identifies resonant excitation bands linked to road spectral content, while Kernel Density Estimation (KDE) maps the probability distribution of displacement states to distinguish stable from variable regimes. The Lyapunov stability assessment and bifurcation analysis reveal critical velocity thresholds and parameter regions marking transitions from stable operation to chaotic motion. The study quantifies the influence of the gravity–damping ratio, mass–damping coupling, control torque ratio, and vertical excitation on dynamic stability. The results provide a methodology for designing stability control systems that ensure safe and comfortable Segway operation across diverse terrains.

The 1.3GHz LCLS-II Cryomodule (CM) has 8 cavities and each cavity has 9 RF cells and 80W heater. The 1.3GHz LCLS-II CM is 12.2-meter-long and contains a total 241.3 liters (equivalent liquid) of liquid helium, which is filling up by the 4KW@2.0K cryogenic plant at the Fermilab Cryomodule Test Facility (CMTF). The dynamic synthesis 2K cryogenic control system is built to control the long time-lag, unstable CM cryogenic system, under different modes operations. A report on the results obtained from the dynamic synthesis long time-lag control in the CMTF 1.3GHz CM test will be presented.