In this research, we propose a robust state estimation framework for a fractional-order Rössler system by implementing Fractional Model Predictive Control (FMPC) with Kalman filtering. The Grünwald–Letnikov (GL) characterization is employed to model the fractional dynamics due to its computational efficiency and ease of implementation, distinguishing it from the more commonly used Caputo and Riemann–Liouville characterizations. Our FMPC framework directly incorporates fractional-order models into the predictive control structure, thereby enhancing control precision and stability. Two scenarios are investigated: one where the state z is measured to infer x and y, and another where x and y are measured to estimate z. In this scenario, realistic conditions with Gaussian noise are considered for the control design. By leveraging the memory properties of fractional derivatives and employing a tailored Extended Kalman Filter (EKF) to ensure state retrieval, our simulations demonstrate that the integrated FMPC–EKF strategy reliably tracks and estimates all state variables under various noise profiles. These results underscore significant improvements in the robustness and accuracy of state estimation for nonlinear, chaotic systems, offering promising implications for practical applications in complex dynamic environments.

In this paper we propose an autoencoder-based End-to-End (E2E) Deep Learning for symbol decoding in fiber optical communications systems. The autoencoder was trained with data encoded in two different ways. First, one-hot encoding which corresponds to the most common approach to train the network in order to optimize the decision regions of each symbol. Then, we introduce the binary coding where the bit sequences that represent the symbols, are positioned in such a way to minimize the error and compensate the distortions introduced by the optical channel. The focus of the study are the generated constellations, symbol decision regions and symbol bit labeling. The autoencoder contributes to better visualization and analysis of the resulting decision regions for scenarios with high and low nonlinearities. It improves the interpretability of the symbol decoding. To the best of our knowledge, this is the first time when a ML-based approach (i.e the autoencoder) estimates not only the ideal locations of the decoded symbols in various constellations (4, 16, 64) but also optimizes the geometrical decision regions of the symbols in the constellation.

This paper describes the identification, robust design and experimental evaluation of the control system of a two-wheeled robot implementing a µ-controller. The design utilizes a single-input multi-output uncertainty model obtained by an appropriate identification procedure. The uncertain model is used in the design of a µ-controller which guarantees robust closed-loop stability and performance. The controller is embedded in a Raspberry Pi 4B microcontroller and the results from the experiments confirm the design.