In this paper, we discuss an improved version of the conventional PID (Proportional–Integral–Derivative) controller, the Dynamically Updated PID (DUPID) controller. The DUPID is a control solution which preserves the advantages of the PID controller and tends to improve them by introducing a quadratic error model in the PID control structure. The quadratic error model is constructed over a window of past error points. The objective is to use the model to give the conventional PID controller the awareness needed to battle the effects caused by the variation of the parameters. The quality of the predictions that the model is able to deliver depends on the appropriate selection of data used for its construction. In this regard, the paper discusses two algorithms, named 1D (one dimensional) and 2D (two dimensional) DUPID. Appropriate to their names, the former selects data based on one coordinate, whereas the latter selects the data based on two coordinates. Both these versions of the DUPID controller are compared to the conventional PID controller with respect to their capabilities of controlling a Continuous Stirred Tank Reactor (CSTR) system with varying parameters in three different scenarios. As a quantifying measure of the control performance, the integral of absolute error (IAE) metric is used. The results from the performed simulations indicated that the two versions of the DUPID controller improved the control performance of the conventional PID controller in all scenarios.
With the development of technology allowing for a rapid expansion of data science and machine learning in our everyday lives, a significant gap is forming in the global job market where the demand for qualified workers in these fields cannot be properly satisfied. This worrying trend calls for an immediate action in education, where these skills must be taught to students at all levels in an efficient and up-to-date manner. This paper gives an overview of the current state of data science and machine learning education globally and both at the high school and university levels, while outlining some illustrative and positive examples. Special focus is given to vocational education and training (VET), where the teaching of these skills is at its very beginning. Also presented and analysed are survey results concerning VET students in Slovenia, Serbia, and North Macedonia, and their knowledge, interests, and prerequisites regarding data science and machine learning. These results confirm the need for development of efficient and accessible curricula and courses on these subjects in vocational schools.
9 April 2018 in IEEE Transactions on Information Forensics and Security
There is an increasing need for everyday communications to be both secure and resistant to external perturbations. We have, therefore, created an experimental implementation of the coupling-function-based secure communication protocol in order to assess its robustness to channel noise. The transmitter and receiver are implemented on single-board computers, thereby facilitating communication of the analog electronic signals. The information signals are encrypted at the transmitter as the time-variability of nonlinear coupling functions between dynamical systems. This results in a complicated nonlinear mixing and scrambling of the information. To replicate the channel noise, analog white noise is added to the encrypted signals. After digitization at the receiver, the decryption is performed with dynamical Bayesian inference to take account of time-varying dynamics in the presence of noise. The dynamical Bayesian approach effectively separates the deterministic information signals from the perturbations of dynamical channel noise. The experimental realization has demonstrated the feasibility, and established the performance, of the protocol for secure, reliable, communication even with high levels of channel noise.
Objective. To determine whether temporal embeddings can recover the discriminative information lost when reducing surface electromyography (sEMG) sensing from dense electrode arrays to a sparse two-channel configuration, and under what conditions they can be integrated without degrading spatial features. Approach. Using a publicly available dataset of 8 subjects performing 10 dynamic finger gestures, we develop a hybrid Transformer optimized for two-channel sEMG. We identify a failure mode in standard additive integration, where the two branches differ in latent magnitude by a factor of four, so the spatial branch dominates the sum. We therefore integrate Time2Vec embeddings through a normalized additive fusion strategy that layer-normalizes both latent distributions before integration. A two-stage curriculum (augmentation-driven pre-training followed by clean-signal fine-tuning) supports robust feature extraction under data scarcity. Main results. Under leave-one-subject-out cross-validation the proposed model achieves a mean F 1-score of 95.9% ± 0.15%, outperforming standard additive, gated, and cross-attention fusion. A gain sweep and a parameter-free normalization variant identify scale alignment as the operative mechanism, as a learnable basis performs worse than fixed sinusoidal encodings under standard addition but exceeds them once normalized. Zero-shot transfer to unseen subjects yields 21.0% ± 2.98%, and classical hand-crafted-feature models collapse comparably, consistent with a sensing constraint rather than a representational one. Supervised calibration with two trials per gesture recovers performance to 96.9% ± 0.52%. Inference runs in 21.5 ms on a consumer CPU core. Significance. Temporal embeddings can compensate for reduced spatial sensing density provided their integration preserves the scale of both representations, enabling rapidly personalizable myoelectric interfaces.