Varela Aldas, José
Thumbnail Image
Preferred name
Varela Aldas, José
Main Affiliation
Ambato
Email
josevarela@uti.edu.ec
ORCID
Scopus Author ID
57192546163

Research Output

Filters
Reset filters

Settings
Now showing 1 - 10 of 136
No Thumbnail Available
Publication

Combining Image Classification and Unmanned Aerial Vehicles to Estimate the State of Explorer Roses

2024 , David Herrera , Pedro Escudero-Villa , Eduardo Cárdenas , Marcelo Ortiz , Varela Aldas, José

The production of Explorer roses has historically been attractive due to the acceptance of the product around the world. This species of roses presents high sensitivity to physical contact and manipulation, creating a challenge to keep the final product quality after cultivation. In this work, we present a system that combines the capabilities of intelligent computer vision and unmanned aerial vehicles (UAVs) to identify the state of roses ready for cultivation. The system uses a deep learning-based approach to estimate Explorer rose crop yields by identifying open and closed rosebuds in the field using videos captured by UAVs. The methodology employs YOLO version 5, along with DeepSORT algorithms and a Kalman filter, to enhance counting precision. The evaluation of the system gave a mean average precision (mAP) of 94.1% on the test dataset, and the rosebud counting results obtained through this technique exhibited a strong correlation (R2 = 0.998) with manual counting. This high accuracy allows one to minimize the manipulation and times used for the tracking and cultivation process.

View
No Thumbnail Available
Publication

Estimation of Unmodeled Dynamics: Nonlinear MPC and Adaptive Control Law With Momentum Observer Dynamic

2024 , Bryan S. Guevara , Luis F. Recalde , Viviana Moya , Varela Aldas, José , Daniel C. Gandolfo , Juan M. Toibero

This article proposes an enhancement to estimate unmodeled dynamics within the simplified dynamic model of a quadcopter by integrating three key methodologies: Nonlinear Model Predictive Control (NMPC), a Momentum Observer Dynamics (MOD), and an adaptive control law. Termed as Adaptive NMPC with MOD, this integrated approach leverages NMPC, implemented using the CasADi framework, for real-time decision-making, while the momentum observer facilitates system state estimation and uncertainty mitigation. Simultaneously, the adaptive control law adjusts parameters to estimate errors in unmodeled dynamics. Through digital twin and Model in Loop (MiL) simulations, the effectiveness of this framework is demonstrated. Specifically, the study focuses on the simplified quadcopter model, acknowledging often overlooked inherent dynamics resulting from the simplification by not considering the nonlinearities induced by the drone's attitude angles. Addressing these unmodeled dynamics is critical, and the Adaptive NMPC with MOD method emerges as a robust solution, showcasing its potential across various scenarios.

View
No Thumbnail Available
Publication

Inverse kinematics of a redundant manipulator robot using constrained optimization

2020 , Varela Aldas, José , Ayala-Chauvin, Manuel Ignacio , Andaluz, V.H. , Santamaría, M.

Redundant manipulative robots are characterized by greater manipulability improving performance but complicating inverse kinematics, on the other hand, optimization techniques allow solving complex problems in robotics applications with greater efficiency. This paper presents the inverse kinematics of a redundant manipulative robot with four degrees of freedom to track a desired trajectory, and considering constraint in manipulability. The optimization problem is proposed using the quadratic position errors of the operative end and the constraint is established by a manipulability index, for this the kinematic model of the robot is determined. The results show the points of singularity of the robot and the performance of the proposal implemented, observing the positional errors and the manipulability for each point of the trajectory. In addition, the optimization is evaluated for two desired manipulability values. Finally, it is concluded that the implemented method optimizes the inverse kinematics to track the desired path while constraining the manipulability. © Springer Nature Switzerland AG 2020.

View
No Thumbnail Available
Publication

3D Object Reconstruction Using Concatenated Matrices with MS Kinect: A Contribution to Interiors Architecture

2020 , Buele, Jorge , Varela Aldas, José , Castellanos E.X. , Jadán Guerrero, Janio , Barberán J.

Interior architecture is part of the individual, social and business life of the human being; it allows structuring the spaces to inhabit, study or work. This document presents the design and implementation of a system that allows the three-dimensional reconstruction of objects with a reduced economic investment. The image acquisition process and treatment of the information with mathematical support that it entails are described. The system involves an MS Kinect as a tool to create a radar that operates with the structured light principle to capture objects at a distance of less than 2 meters. The development of the scripts is done in the MATLAB software and in the same way the graphical interface that is presented to the user. As part of the initial tests of this prototype, the digitization of geometric shape structures has been performed with an accuracy of over 98%. This validates its efficient operation, which serves as the basis for the development of modeling in interior architecture for future work. © 2020, Springer Nature Switzerland AG.

View
No Thumbnail Available
Publication

Using LSTM Autoencoder and Dynamic Signal Decomposition for Efficient PAPR Reduction in UWA-OFDM

2025 , Mansoor Jan , Syed Agha Hassnain Mohsan , Víctor P. Gil Jiménez , Muhammad Aman , Samih M. Mostafa , Varela Aldas, José

Underwater Acoustic Communication (UWA) is an emerging wireless technology that enables reliable and long range data transmission in challenging underwater environments. To support high data rates and mitigate Inter Symbol interference (ISI) caused by multipath propagation, Orthogonal Frequency Division Multiplexing (OFDM) is widely adopted in UWA systems. High Peak-to-Average Power Ratio (PAPR) continues to be a significant constraint in OFDM based UWA systems, as it causes nonlinear distortion in power-constrained acoustic transmitters and results in spectrum regrowth. This study presents a data driven strategy for PAPR reduction using a Long Short Term Memory-Autoencoder (LSTM-AE) as well as noise reduction from time-domain OFDM signals using the Local Mean Decomposition technique (LMD). The LSTM-AE is designed to learn the basic components of the OFDM waveform and rebuild a low PAPR variant without sacrificing signal integrity. Simulation results indicate that the suggested strategy significantly decreases PAPR while maintaining Bit Error Rate (BER) performance across different Signal-to-Noise Ratio (SNR) levels. Moreover, Complementary Cumulative Distribution Function (CCDF) analysis demonstrates substantial improvement relative to conventional PAPR reduction methods, including Tone Reservation (TR), Partial Transmit Sequence (PTS), Active Constellation Extension (ACE) and Fully Connected Neural Network (FCNN). Compared to the original OFDM signal, the LSTM-AE approach reduced PAPR by 9.4 dB, reaching 4.8 dB. It also maintain better energy efficiency than ACE, FCNN, TR, and PTS at all SNR levels. The suggested LSTM-AE method is a feasible low complexity alternative for improving the efficiency and resilience of UWA OFDM systems

View
No Thumbnail Available
Publication

Detection System for Domestic Environmental Pollutants Based on ThingSpeak

2023 , Varela Aldas, José , Miranda M. , Lasluisa G.

Internet of things platforms is essential components for project development with connectivity. Thus, these have positioned themselves in the market by offering multiple cloud services, such as a database, application programming interface, and more. This paper presents a system that collects data from air concentration sensors to detect the presence of pollutants. The electronic circuit uses a system-on-chip with WiFi connectivity and the MQ2 and MQ3 sensors, which detect liquefied petroleum gas and alcohol concentrations in the air, respectively. The system is based on the ThingSpeak platform, which receives and graphs data to see readings outside expected ranges. The results present the concentration levels of the MQ2 and MQ3 sensors, indicating the correct functioning of the system, and future applications of the proposed design are discussed. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

View
No Thumbnail Available
Publication

Mobile Manipulator for Hospital Care Using Firebase

2022 , Varela Aldas, José , Buele, Jorge , Guerrero-Núñez S , Andaluz V.H.

The COVID-19 pandemic has shown that the use of the technology in medicine is no longer a luxury, but a necessity. The use of the robotics in the treatment of diseases and physical therapies is limited in Latin America due to the high acquisition and maintenance costs. This document proposes the design, development, and evaluation of a robotic system for the guided monitoring of patients, through remote control using a mobile application. Within the methodology, four phases were proposed: planning, design, development, and evaluation. The 3D design is done using the Tinkercad software, which facilitates the construction of the pieces using 3D printing technology. The ESP32 board is the main element that receives the signals from the sensors and controls the actions of the actuators through the orders received from Firebase. For the development of the application, App inventor is used, building a friendly and easy-to-use interface. To validate this proposal, experimental tests were carried out with two patients in a medical center. In addition, a parameter compliance questionnaire was applied to the robot, obtaining a score of 92.6%, and the mobile application obtained 72.5% in the usability test. All this confirms an efficient care proposal, with a reduced investment. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

View
No Thumbnail Available
Publication

Voltammetric Electronic Tongues Applied to Classify Sucrose Samples Through Multivariate Analysis

2021 , Fuentes, E.M. , Varela Aldas, José , Verdú S. , Meló R.G. , Alcañiz M.

The aim of the present study was to classify samples of sugar with different concentrations through a Voltammetric Electronic tongues (VET), with a generic pulse sequence consisted of 22 pulses ranging from –1000 mV to + 1000 mV with a duration of 20 ms/pulse over different samples such as 1.25mM, 2.5mM, 5mM and 10mM, of sucrose concentration, these were measured 4 times each concentration and the test was developed 4 times, giving a total number of 506.880 data supervised learning algorithm using support vector machine was employed, choosing a linear function as a classifying element. In the training, 75% of the data was used to determine the coefficients of the classification function, and the remaining (25%) was used to evaluate the performance of the proposal. The results showed a concordance of more than 80% in the separation of sample, allowing to conclude as acceptable the performance of the classifier and the data acquired through the voltammetric tongue. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

View
No Thumbnail Available
Publication

IoT-Based User Interface for Remote Control of a Mobile Robot

2023 , Varela Aldas, José , Palacios-Navarro G.

Recent advancements in mobile robot research have resulted in the development of precise robot control tools, while information technology research has focused on the Internet of Things (IoT) in the context of the fourth industrial revolution. This study evaluates a user interface designed for remote control of the Crowbot BOLT robot. This robot utilizes the ESP32 board and is controlled through the M5Stack Core2 kit with a touch screen. The user interface offers two modes of operation: touch-based buttons for movement control and gyroscope control based on the M5Stack’s angular position. Communication between the robot and the user interface is established using the MQTT protocol through the ThingSpeak server, allowing control from any location with a line of sight and internet connectivity. Operation data is collected by recording control orders and measuring sending times, while user acceptance is evaluated using an IoT-based technology acceptance model. The results indicate the need for remote control response time improvement and reveal low scores in perceived usefulness and influence social. In conclusion, the study demonstrates the feasibility of remote control of a mobile robot using the MQTT protocol, providing valuable insights for similar applications and considering user recommendations for future enhancements and system expansion. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

View
No Thumbnail Available
Publication

Enhancing feature learning of hyperspectral imaging using shallow autoencoder by adding parallel paths encoding

2025 , Bibi Noor Asmat , Hafiz Syed Muhammad Bilal , M. Irfan Uddin , Faten Khalid Karim , Samih M. Mostafa , Varela Aldas, José

Conventional image formats have limited information conveyance, while Hyperspectral Imaging (HSI) offers a broader representation through continuous spectral bands, capturing hundreds of spectral features. However, this abundance leads to redundant information, posing a computational challenge for deep learning models. Thus, models must effectively extract indicative features. HSI’s non-linear nature, influenced by environmental factors, necessitates both linear and non-linear modeling techniques for feature extraction. While PCA and ICA, being linear methods, may overlook complex patterns, Autoencoders (AE) can capture and represent non-linear features. Yet, AEs can be biased by unbalanced datasets, emphasizing majority class features and neglecting minority class characteristics, highlighting the need for careful dataset preparation. To address this, the Dual-Path AE (D-Path-AE) model has been proposed, which enhances non-linear feature acquisition through concurrent encoding pathways. This model also employs a down-sampling strategy to reduce bias towards majority classes. The study compared the efficacy of dimensionality reduction using the Naïve Autoencoder (Naïve AE) and D-Path-AE. Classification capabilities were assessed using Decision Tree, Support Vector Machine, and K-Nearest Neighbors (KNN) classifiers on datasets from Pavia Center, Salinas, and Kennedy Space Center. Results demonstrate that the D-Path-AE outperforms both linear dimensionality reduction models and Naïve AE, achieving an Overall Accuracy of up to 98.31% on the Pavia Center dataset using the KNN classifier, indicating superior classification capabilities.

View