German Alberto Echaiz Espinoza

Surface electromyography (sEMG) is a critical tool for biomedical research and rehabilitation; however, commercial multi-channel systems exceed $400~USD, employ proprietary firmware, and limit scientific reproducibility. This paper presents MyoScope μPy, a fully open-source hardware system that overcomes these barriers: it provides four-channel sEMG acquisition using passive stainless-steel dry electrodes for less than $46 USD, an order of magnitude below the closest commercial analog. The analog front-end integrates an INA828 instrumentation amplifier (G ≈ 311 V/V), zero-drift active filters (OPA333, AD8629), and a 60 Hz Twin-T notch filter that suppresses power-line interference down to the noise floor, even with floating electrodes in ungrounded environments. The firmware, implemented in MicroPython on a Pyboard v1.1, achieves deterministic sampling at 2500 Hz per channel through a zero-allocation triple-buffering architecture, demonstrating that a high-level language can meet real-time requirements in biomedical instrumentation. Experimental validation across five progressive scenarios confirms an input-referred noise of 3.51 μVrms (7.96 effective bits) on the best-performing channel and a functional SNR of up to 19.4 dB (rest vs maximum voluntary contraction). With exclusive LiPo battery power (autonomy > 5h), complete galvanic isolation, and all design files released under a CC BY 4.0 license, MyoScope μPy constitutes a readily reproducible platform for budget-constrained laboratories, educational settings, and myoelectric control research projects.

The control of drum-boiler systems is challenging due to their highly nonlinear dynamics and shrink-and-swell phenomena that complicate level regulation. To this end, a cascade control strategy with feedback action is designed to reduce the effects of contraction and expansion within the boiler, as well as to reduce the effect of disturbances under variable load conditions. The simulation results demonstrate the superiority and reliability of the three-element cascade controller compared to the conventional single-element scheme, achieving faster stability, reducing oscillations, and improving tracking performance. Unlike previous work, this study provides complete mathematical derivations of the fundamental equations and a complete and replicable implementation in MATLAB/Simulink, ensuring both transparency and industrial relevance. The developed simulator provides a robust platform for further research, operator training, and the design of advanced control strategies in thermal power plants.

Controlling nonlinear thermal systems like rotary cement kilns is a long-standing problem in industrial automation since they are dynamic and hard to predict. Fuzzy Logic Controllers (FLCs) present a strong alternative; nonetheless, the essential defuzzification phase - responsible for transforming fuzzy outputs into actionable values - has been inadequately examined in comparative analyses. This project creates a Mamdani-type FLC with 88 rules based on experts and four main inputs: combustion temperature, furnace torque, CO percentage, and preheater temperature. It will be used to control a high-capacity rotary kiln. We comprehensively examine five defuzzification methods: centroid, bisector, Smallest of Maximum (SOM), Middle of Maximum (MOM), and Largest of Maximum (LOM), using simulations generated in MATLAB’s Fuzzy Logic Toolbox. The results reveal that the output is generally consistent among approaches, except for the kiln feed rate, which is quite sensitive. The research illustrates that the selection of defuzzification strategy significantly influences control performance, with maxima-based approaches providing enhanced stability. These results provide practical advice for developing strong fuzzy controllers for complicated industrial processes.

Most autonomous navigation approaches for mobile robots rely either on 3D LiDAR sensors, which are often costly, or on horizontally mounted 2D LiDAR sensors, which do not support threedimensional mapping. This paper presents the implementation of a 3D Simultaneous Localization and Mapping system integrated with autonomous navigation in a mobile robot using a Translational Single LiDAR 2D (TSL-2D) configuration: a three-dimensional mapping mechanism based on a single vertically mounted 2D LiDAR sensor. To verify the feasibility of the proposed approach, the methodology combines the mathematical foundations, implementation and evaluation of global planning algorithms—Dijkstra, A-star (A*), and Rapidly-exploring Random Tree Star (RRT*)—within the Robot Operating System 2 (ROS 2) framework, integrating point cloud generation from vertical LiDAR sweeps. Experiments are conducted both in simulated environments using Gazebo and in real-world scenarios. The qualitative results demonstrate that the system is reliable for evaluating autonomous navigation algorithms, including local and motion planners. Quantitatively, as a specific outcome of the proposed system, RRT* produces smoother and shorter trajectories compared to A* and Dijkstra, achieving a 7.62% reduction in path length. Furthermore, in comparison to RGB-D camera-based systems reported in the literature, the TSL-2D mechanism delivers superior precision in generating 2D projected maps from the acquired point clouds.

The use of recurrent neural networks has proven effective in time series prediction tasks such as weather. However, their use in resource-limited systems such as MCUs presents difficulties in terms of both size and stability with longer prediction windows. In this context, we propose a variant of the LSTM model, which we call SE-LSTM (Single Embedding LSTM), which uses embedding techniques to vectorially represent seasonality and latent patterns through variables such as temperature and humidity. The proposal is systematically compared in two parts: The first compares it against other reference architectures such as CNN-LSTM, TCN, LMU, and TPA-LSTM. The second stage, which includes implementation, compares it against the CNN-LSTM, LSTM, and TCN networks. Metrics such as the MAE and MSE are used along with the network weight, a crucial aspect for MCUs such as an ESP32 or Raspberry Pi Pico. An analysis of the memory usage, energy consumption, and generalization across different regions is also included. The results show that the use of embedding optimizes the network space without sacrificing the performance, which is crucial for edge computing. This research is part of a larger project, which focuses on improving agricultural monitoring systems.

This paper presents the analysis and simulation of synchronization between an active power grid and a power generation system driven by a Francis-type hydraulic turbine, using the model available in Simulink. Synchronization is a fundamental process in electrical systems, allowing the safe and efficient connection of generators in parallel with the grid. To achieve this, it is necessary to adjust the generator’s voltage, frequency, and phase angle to match the grid values. Although physical equipment exists for this interconnection, this research simulates a soft synchronizer as an alternative to optimize the synchronization angle adjustment. To analyze the voltage signals, the d-q-0 reference frame transformation is employed, simplifying processing and enabling more precise detection of variations in voltage magnitude and phase angle. The theoretical framework of this research encompasses power system theory, electrical machines, and control, as well as mathematical models of Francis hydraulic turbines. Using Simulink libraries in MATLAB, the results obtained are evaluated based on variations in voltage, frequency, and phase angle during the transient synchronization period. The results show a considerable decrease in generator magnitude oscillations.

This paper presents a computational simulation of a stainless steel straight-tube Coriolis sensor, focusing on the design and validation of a mathematical model for mass flow measurement. The Finite Element Method (FEM), implemented in GNU Octave, is used to solve the differential equations governing the tube’s vibrational behavior. The simulation determines the angular resonance frequency (723.6 rad/s) and evaluates mass flow for fluid velocities ranging from 0.5 m/s to 25 m/s. The results are validated against reference data from existing literature, demonstrating strong agreement and confirming the accuracy of the proposed model. These findings reinforce the potential of Coriolis sensors for precise and efficient mass flow measurement in industrial applications such as chemical processing and fluid transport monitoring.

Households should use good and efficient energy management practices to minimize energy consumption and maximize the use of renewable energy sources. This paper describes a novel approach to regulating household energy consumption that combines solar panels and machine learning approaches. The suggested system estimates energy demand and solar power generation with high accuracy using past data on energy consumption and weather forecasts. The machine learning model dynamically adjusts energy usage patterns and storage solutions, thereby maximizing solar energy utilization and minimizing grid dependency. In addition, the local grid tariff cost is compared to determine the time required in which the implemented system becomes self-sufficient. Simulation results demonstrate significant improvements in energy efficiency and cost savings for residential users.

Food shortages in Peru are frequently exacerbated by roadblocks due to social protests and seasonal variations, restrict food production, and intensify shortages, particularly in regions with limited cultivable land. To address this challenge, an automated control system and IoT architecture for wireless monitoring were developed to replicate specific climatic conditions within greenhouses, allowing the cultivation of crops typically restricted to certain regions and providing a potential solution to improve agricultural productivity and mitigate food scarcity. The system regulates critical environmental parameters such as temperature, irrigation and luminosity, while wireless monitoring of temperature, water level, luminosity, and pH is achieved through RF communication, with data transmitted to a database via Wi-Fi and visualized in real-time using a cross platform application. In addition, the NRF24L01 radio frequency communication modules were optimal for the application due to their ease of integration with the Arduino UNO development board, affordability, range, and market availability. The system was validated in a pilot-scale greenhouse. Further research is needed to evaluate its long-term scalability and efficiency in diverse agricultural settings, as these aspects remain critical to its broader applicability and impact.