Rehabilitation robotics is an important field of study focused on improving the gait rehabilitation of people affected by neurological disorders, ictus, cerebral palsy, and spinal cord injuries, among others. The study of rehabilitation robotics includes medical activities, kinematics dynamics, and control analysis. This book presents a complete and exhaustive analysis of the kinematics and dynamics of exoskeleton robots for rehabilitation. The forward and inverse kinematics are studied using the geometric, Denavit-Hartenberg, and screw theory approach. The dynamics analysis of exoskeleton robots using Newton-Euler, Euler–Lagrange, and D'Alembert formulation are also studied. Moreover, the main control techniques for exoskeleton robots are analyzed, including robust control, impedance control, adaptive control, Lyapunov functions, and uncertainties found in dynamic systems. The book includes MATLAB applications and examples.

This chapter presents a comprehensive review of lower limb exoskeleton robots for rehabilitation. The most important features of each exoskeleton are discussed. In addition, a review of the applications of the exoskeletons according to different pathologies is presented. Then, challenges and issues in rehabilitation robotics are discussed as well as the exoskeleton market, sales, and forecast. This work covers different aspects of the development and current trends in robotic exoskeletons, providing as a complete review of the state-of-the-art of these devices.

In this book chapter is evinced the theoretical and experimental results regarding the adaptive control of rehabilitation robotics. In the first subsections it is explained the basics of adaptive control for different kinds of robots and mechanical systems. Meanwhile in the second part of this chapter is evinced three adaptive control strategies for rehabilitation robotics. These strategies are based on proportional-integral-derivative, sliding mode control, and backstepping adaptive control strategies. So in this way, it will be covered the main theoretical background for the adaptive control of rehabilitation robots. The ALICE robotics mechanism will be analyzed in this chapter in order to verify the stability and error convergence to zero in finite time for trajectory tracking purposes. The Lyapunov theorem is crucial to design these adaptive controllers, in order to meet the previous mentioned closed loop system requirements.

Due to the diverse kinds of uncertainties and disturbances found in rehabilitation robots, this chapter is devoted to evince diverse kinds of impedance controllers for these kinds of mechanisms. Impedance controllers consist in considering diverse kinds of environments in which the rehabilitation robot operates. These kinds of environments disturbances are generally found in the forces and torques that the patients generate for diverse kinds of mechanisms. The first step to design these kinds of mechanisms is to design a “mechanical circuit” (similar to an electrical circuit) with the equivalent circuit elements like inertia, mass, stiffness coefficient, etc. Then, as seen later, the equivalent Thevenin or Norton equivalent circuits are designed taking into consideration if the circuit is inertial, capacitive, or resistive. In this chapter, two numerical examples are shown, implementing the respective impedance controller. The first numerical example consists of the design of an impedance controller for a five bar linkage, and the second numerical example consists in designing an appropriated controller for a gait robot. Exercise to be solved appears at the end of the chapter.

In this chapter it is explained the construction of robust control Lyapunov functions. This chapter begins with the definition of robust stability and robust performance for linear and nonlinear dynamic systems in order to obtain the robust requirements to measure this index in order to ensure the best performance of the closed loop robust system. Then the construction of robust control for linear and nonlinear systems is explained in order to define the theoretical conditions for the robust control system design. One of the main contributions of this book chapter is that the robust stability conditions and robust performance are extended to the discrete time case, something that is very limited reported in the literature and that this book chapter provides a novel contribution to the field. Besides, the derivation of control laws for the continuous and discrete time cases is presented, in order to provide the reader the mathematical background to design controllers for this kind of systems. Another important contribution of this book chapter is that the derivation of optimal robust control laws is presented, in specific the design of optimal robust linear quadratic regulators for the continuous time case. Finally, illustrative examples are presented to elucidate the experimental results of this research study in order to obtain the optimal closed loop system performance.

In this chapter two robust control strategies for nonlinear dynamic systems are presented. This chapter begins with a brief review of the most common kind of uncertainties found nowadays by considering their mathematical and topological foundations. The kinds of uncertainties explained in this chapter are basically parametric, nonparametric, matched and mismatched uncertainties describing the mathematical properties, in specific, the topological properties taking into consideration that recently there are a plenty of mathematical representations due to the vast amount of uncertainties found in nonlinear and other kinds of complex dynamical systems. One important result explained in this chapter is that the well-known robust control Lyapunov functions (RCLFs) are explained by taking into consideration that the closed loop robust stability and the robust control law are found by an appropriate construction of the RCLF. Besides in this chapter is explained the theoretical background of the design of robust controllers by considering uncertainties in nonlinear dynamic systems, and by implementing the respective Lyapunov functional a suitable robust controller can be obtained in order to ensure robust stability of a closed loop system. Besides, the design of optimal robust controllers is explained by taking into consideration that a performance functional along with the respective Hamiltonian is shown in order to design the robust optimal control law by taking into consideration the uncertainties in the system. One of the main contributions of this chapter is that a robust controller is designed for systems in the Euler-Lagrange formulation by designing the Lagrangian in order to obtain the required formulation. The mass and inertia uncertainties are considered in this study with the respective topological definition, and by using an appropriate Lyapunov functional, the robust control law is found. This control law can be implemented in a variety of mechanical and robotic systems for trajectory tracking purposes. Finally, an optimal robust controller for mechanical systems is presented in this chapter by taking into consideration a Lyapunov functional and a performance functional for the robust optimal control law design. This robust optimal controller is implemented in a nonlinear forced mass-spring system, corroborating a high performance.

In this chapter, the dynamics equations of motion of ALICE (Assistive Lower lImb Controlled Exoskeleton) exoskeleton robot are found. First the ALICE project is presented. Then, the D’Alembert formulation in order to find the dynamic equations of motion is applied. Finally, these equations are validated through the joint’s torque obtained with a dynamic simulation using an ALICE’s MATLAB/Simulink equivalent model, and the dataset used for testing was obtained from healthy subjects and was captured with the commercial medical equipment CODA motion system. MATLAB as programming tool and testing platform is used.

In this chapter, the dynamics analysis of ALICE (Assistive Lower lImb Controlled Exoskeleton) Rehabilitation Robot is presented. First the ALICE project is presented. Then, the recursive Newton-Euler approach in order to find the dynamic equations of motion is applied. Finally, the equations are validated comparing the joint’s torque obtained with the dynamic simulation using an ALICE’s MATLAB equivalent model, the dataset using for testing correspond to real gait data obtained from healthy subjects and were captured with the commercial medical equipment CODA motion system. MATLAB as programming tool and testing platform is used.

This chapter presents the fundamentals of exoskeleton robots and their applications. A complete review of the upper and lower limb exoskeletons robots for rehabilitation, for both commercial devices and those in the advanced stage of development, is discussed. The most important characteristics of each exoskeleton and the level of development are presented according to the technology maturity level (TRL) metric proposed by NASA. In addition, it has a review of the applications of each of them according to the types of pathologies. Finally, future trends of rehabilitation robotics are discussed.

This chapter presents the fundamentals of exoskeleton robots and their applications. A complete review of the upper and lower limb exoskeletons robots for rehabilitation, for both commercial devices and those in the advanced stage of development, is discussed. The most important characteristics of each exoskeleton and the level of development are presented according to the technology maturity level (TRL) metric proposed by NASA. In addition, it has a review of the applications of each of them according to the types of pathologies. Finally, future trends of rehabilitation robotics are discussed.

Convolutional neural networks (CNNs) promise to transform modern agriculture by addressing some of the most pressing challenges. This project focuses on developing an effective CNN to automate the selection of rambutans according to maturity and minimize losses caused by premature harvesting and inclusion of low-quality fruit. As well as identifying the main limitations and challenges that arise when developing a CNN focused on this sorting process. For this purpose, a database of images captured in Honduran rambutan production farms was obtained. The database was composed of 1,540 source images that were transformed to 10,144 after Aumentación. Six object detection Modelos focused on fruit classification were then created. The best result achieved a mAP of 64.1 %, a precision of 77.2 % and a recall of 58.6 %. Finally, the main limitations for the development of such a network were identified. These were divided into the categories of dataset availability and management, data expansion and resource limitations and specific detection challenges.

This paper aims to develop a numerical method to solve the Schrodinger equation in a time-dependent dimension for mobility problem in a N-MOSFET using the finite difference method. The proposed method is implemented using the GNU Octave tool and the fsolve function to solve nonlinear algebraic equations. This method is expected to provide an accurate and efficient solution to the mobility problem in a N-MOSFET, which could have applications in fields such as quantum physics and electronic engineering.

This article provides an analysis of the state of the art of the STEAM approach in education, the different emergent methodologies that accompany this new paradigm, and the disruptive technologies that are being integrated. Then, a case of success of recommended methodologies in a robotics and disruptive technologies academy is presented. Finally, the challenges that educational institutions must face to effectively integrate STEAM education are addressed.

Parallel mechanism has a great advantage over serial mechanism, this is why day by day in the industry many of them are being made for different applications that require speed, precision and strength. Is known that synthesis of these mechanism is more complex than serial mechanism, for this reason, specifications are sought for its design that can facilitate its analysis. This document provides an analytical synthesis of a five-bar mechanism and a methodology to determine the size of the links and then check their mobility using the Grashoff criterion, furthermore, the workspace of the mechanism is calculated to check that the final effector of the mechanism arrives to the point of interest. Finally, the design of the workspace is presented in Geogebra Software as a simulation, including the final mechanism that was manufactured.

This article provides a prototype analysis for Parkin-son's disease; focused on the anomalies on the deterioration of nerve cells, which produce rhythmic shaking (tremors), generally a warning sign for a diagnosis of Parkinson's.

This study investigates the mechanical properties of polyester resin composites reinforced with sisal fibers in filament and staple forms to assess their viability as sustainable alternatives to traditional materials. Tensile tests were conducted following ASTM standards to evaluate the mechanical performance and sustainability of sisal fiber composites for potential applications in various industrial sectors, such as automotive and aerospace. The discussion addressed how sisal fibers could replace conventional materials in a sustainable wav

This paper describes the creation of the second prototype of a device that helps blind people access new technologies. The whole project is named “Optical Cane,” which is an embedded system integrating an application designed on Flutter and programmed on Visual Studio Code and Thonny. Additionally, it includes an adaptable device created with 3D modeling, an ESP32 (taking advantage of its easy access to the IoT of the shield), and multiple sensors programmed with Micro-Python. These sensors detect objects and analyze them to provide the highest possible accuracy in measuring obstacles, either through the app or by giving a tactile or sound signal. Incorporating different sensors such as ultrasonic, camera, and gyroscopic sensors helps create a rich environment, providing a friendly experience for users with visual disabilities. This prototype demonstrates that accessibility and technology can become even more affordable for people who want to enhance their perception of the world.

This paper presents the results obtained from the PayPal integration project using PHP, JavaScript, and MySQL. The purpose was to implement improvements in the purchasing processes for small and medium-sized enterprises that are in a growth phase, in order to guide them towards an approach based on audits and security. The development of this project yielded positive results and benefits, which include facilitating response times, reducing lost sales, and improving control over the management of customer data and purchase orders. In a globalized context with significant technological advancements, small and medium-sized enterprises cannot afford to fall behind. Therefore, the proposed project allowed the use of available technologies and provided access to them in a practical and more economically feasible way than the complete design of software. Additionally, improvements and adaptability points were validated through testing, according to the conditions.

The Coffee Berry Borer (CBB), Hypothenemus hampei, is one of the most serious threats to global coffee production. This study provides a thorough analysis of the different types of traps used for detecting and controlling the CBB, outlines the development of a preliminary electronic device capable of capturing images of the pest in coffee plantations, and details the creation of a dedicated database. Through a systematic literature review of 13 relevant studies, we examined various trap designs and attractant combinations used for CBB detection. Additionally, we developed a prototype device equipped with an ESP32-CAM microcontroller and solar power capability for continuous field monitoring. The device captures high-resolution images at 15-minute intervals. A preliminary database of 500 images was created. Our findings suggest that effective trap placement typically ranges from 1.2 to 1.5 meters above ground, with a 3:1 methanol-ethanol mixture serving as the optimal attractant. This research contributes to the development of automated CBB monitoring systems and establishes a foundation for future applications of artificial intelligence (AI) in pest management strategies.

This study explores how effective VIS-NIR spectroscopy, combined with machine learning and deep learning techniques, is in classifying various apple varieties gathered from two distinct locations. The study analyzed the diffuse reflectance spectra of three types of Chinese apples: the Fuji apple, the Red Star apple, and the Gala apple. Each apple variety was collected in two distinct collection locations: Shandong and Shaanxi. To achieve this, a comparison between machine learning methods (partial least squares discriminant analysis (PLS-DA) and deep learning methods (artificial neural networks discriminant analysis (ANN-DA)) has been performed. The results show that the ANN-DA models outperform the PLS-DA and SVM models. For the entire sample set, the fact of having three different apple varieties makes all three algorithms the task of classifying correctly more difficult. When the two locations are classified using samples from the same variety, ANN has an accuracy in the calibration and prediction sets of 100%, except for the fuji variety, which reaches 94%. These results show that NIR spectroscopy coupled with Machine Learning or Deep Learning models is a capable method for identifying the apple growing region even for industrial-scale adoption.

This study analyzes the technical, economic, and social barriers limiting access to artificial intelligence (AI) in less developed regions. Using a non-experimental, quantitative approach, methods such as interviews and content analysis are employed to understand perceptions and contexts surrounding AI democratization. The research design does not aim to manipulate variables but rather to analyze existing relationships. Only 30% of countries, mainly in North America, Europe, and parts of Asia, have significant access to advanced technologies, while less than 20% have implemented national AI policies. Barriers include a lack of infrastructure and high costs, restricting AI adoption in underdeveloped areas. Additionally, ethical implications, such as privacy, transparency, and fairness, are highlighted. To democratize AI, it is crucial to include diverse communities in its design and establish policies that promote equitable access.

Agriculture plays a fundamental role in Central America’s economy, yet it faces significant challenges that demand attention. This study aims to examine the technologies and research applied within the context of Agriculture 4.0 in the Central American region, evaluating its adoption and providing a comprehensive overview of the integration of emerging technologies. The analysis highlights that Costa Rica leads research in Agriculture 4.0 in the region, followed by Panama, with a particular focus on crops such as coffee and rice. Despite the increase in scientific production, there remains a notable lack of research in emerging technological areas. Improving agriculture in Central America necessitates investment in modern agricultural technology, promotion of sustainable practices, and enhancement of farmers’ skills. The implementation of Agriculture 4.0, leveraging artificial intelligence, IoT, IoUT, UGVs, digital twins, remote sensing, wireless sensor networks, and automation, is crucial for the sector’s advancement. Additionally, it is imperative to promote agricultural scientific research to develop crops resilient to climate change and more efficient production methods.

This article examines how artificial intelligence is reshaping education through personalized learning, administrative automation, and immersive technologies. It highlights innovations such as intelligent tutoring systems, automatic assessment, and VR/AR tools, while addressing opportunities and challenges related to ethics, privacy, and equity. The study underscores AI’s potential to enhance educational effectiveness and adapt teaching methods in STEAM education.

This paper presents a comprehensive analysis of electromobility trends and policies across Central America, focusing on Guatemala, El Salvador, Honduras, Nicaragua, Costa Rica, and Panama. We examine the current state of electric vehicle (EV) adoption, charging infrastructure development, and policy frameworks in each country. The study highlights significant progress in countries like Costa Rica and Panama, which have implemented robust incentive structures and ambitious targets for EV integration. We also discuss challenges faced by the region, including the need for expanded charging networks and regulatory harmonization. The paper explores the role of public-private partnerships in accelerating EV adoption and the importance of renewable energy integration to maximize environmental benefits. Our analysis reveals a growing momentum towards electromobility in Central America, driven by climate commitments, energy security concerns, and economic opportunities. We conclude by identifying key future trends and policy recommendations to further advance electromobility in the region.

This paper presents the development of an Inertial Measurement Unit (IMU)-based gait acquisition system designed for biomechanical analysis. The proposed system uses IMU sensors to capture and analyze gait patterns, providing detailed insights into human movement that are critical for applications such as exoskeleton design and rehabilitation. The data acquisition process is enhanced by the implementation of a Kalman filter, which ensures accurate and reliable processing of sensor data. The suitability of the system for comprehensive biomechanical studies is demonstrated by comparing previously published works, making it a valuable tool for both clinical and research applications in the field of human motion analysis.

The design and development of a robot capable of climbing stairs and moving on uneven terrains represents a significant advancement in the field of robotics, particularly in applications where adaptability in challenging environments is crucial. This type of robot becomes an invaluable tool in a variety of scenarios, from search and rescue operations in disaster-affected areas to exploration tasks in unknown or inaccessible terrains for humans. What sets this robot apart is its efficient rotational capacity across various surfaces. This capability not only allows it to navigate obstacles more effectively but also enhances its practical utility when faced with varied and difficult terrains. Its ability to adapt to different types of surfaces, whether smooth, rough, inclined, or irregular, makes it extremely versatile and suitable for a wide range of applications. Furthermore, the efficiency in the robot's rotational capacity not only enhances its mobility but can also have significant implications in terms of energy consumption and durability. An optimized rotation system can reduce the load on the robot's components and prolong its lifespan, making it a more sustainable and economically viable long-term solution.

Optimization challenges in industrial engineering, particularly in economic order quantity (EOQ) and materials requirement planning (MRP), have traditionally been complex. This research addresses critical limitations in existing production and inventory management models by addressing recent computational advancements. We propose a comprehensive approach to resolving large-scale industrial engineering optimization problems by integrating high-level programming languages and advanced optimization tools. The study focuses on developing a generic Python-based optimization algorithm using a reference optimization model and Gurobi solver, with primary contributions including: (i) systematic exploration of optimization methods in industrial engineering; (ii) development of a flexible, scalable optimization approach; (iii) demonstration of computational techniques' potential in solving complex production planning challenges. By bridging theoretical optimization models with practical implementation, this research offers a cost-effective solution that extends beyond traditional limitations of economic order quantity and production lot sizing methodologies.

This paper analyzes the synergy between nuclear energy and green hydrogen production, highlighting thoriumbased nuclear fission as a stable, low-carbon energy source to power electrolysis processes. It assesses the viability of green hydrogen in hard-to-decarbonize sectors such as the steel industry, ammonia production, heavy transport, and industrial and agricultural machinery. The analysis emphasizes benefits in energy efficiency while also identifying key challenges, including high initial investment, the need for strong regulatory frameworks, and public acceptance of nuclear energy. Finally, future research directions are proposed to validate and optimize this technological alternative.

This study presents a comprehensive kinematic analysis of a five-bar parallel robot, using vector analysis as the primary tool to solve both the forward and inverse kinematics of the mechanism. Through this approach, the geometric relationships and constraints imposed by the robot's links and joints are addressed, allowing for the determination of the end-effector's positions and orientations as functions of the input angles. To validate the obtained results, a comparison is made between the positions calculated using the vector model and those generated through a five-bar parallel robot simulator developed in GeoGebra. Additionally, a Matlab-based software is implemented to simulate the robot's kinematics, providing an additional platform for result verification. Finally, the advantages and limitations of vector analysis are discussed in comparison to other kinematic solution methods, such as numerical and algebraic approaches, highlighting applications and potential areas for improvement in the field of parallel robots.

This article presents a systematic review on the application of Large Language Models in smart manufacturing, highlighting their potential to transform key industrial processes such as predictive maintenance, human-machine interaction, automated documentation, and decision optimization. Recent cases across various domains, including electronic design, industrial programming, and decision support systems, are analyzed, demonstrating the versatility of LLMs in processing large volumes of data, enabling natural communication with operators, and generating proactive recommendations. The review also identifies significant barriers to adoption in Latin America, including limitations in technological infrastructure, insufficient investment, and limited availability of specialized AI personnel. This study emphasizes the need for public policies, professional training, and collaboration among academy, industry, and government to promote the effective and sustainable implementation of LLMs, adapted to local contexts and capable of closing existing gaps in the region.

Artificial Intelligence (AI) has become a transformative tool in education, particularly in STEAM (Science, Technology, Engineering, Arts, and Mathematics) disciplines. This paper examines the integration of AIBOT, an advanced Intelligent Tutoring System (ITS) and adaptive learning platform, within the STEAM Robotics Academy to personalize learning experiences, enhance student engagement, and address diverse educational needs. By leveraging AIBOT’s capabilities, such as step-by-step guidance, real-time feedback, simulation environments, and performance analysis, the study highlights its potential to revolutionize STEAM education. Additionally, the paper explores challenges such as ethical considerations, teacher training, and infrastructure requirements, offering actionable insights for educators, policymakers, and technologists. This analysis contributes to the growing body of knowledge on AI’s role in transforming education and fostering innovation in STEAM fields.

Marine monitoring is essential for the conservation of ecosystems and the sustainable management of marine resources, especially in Latin America, which has extensive and diverse maritime areas. This article presents a systematic review of recent technological advances applied to marine monitoring in the region, covering the Internet of Things (IoT), remote sensing, artificial intelligence, drones, and marine robotics. Using the PRISMA methodology, key technologies, their geographical distribution, and the challenges in their implementation are identified. The results highlight the need to integrate innovative technologies and promote regional collaboration to strengthen marine monitoring systems, supporting the Blue Economy and sustainable environmental management.

The analysis and classification of white blood cells (leukocytes) are critical for diagnosing numerous diseases, as they constitute the immune system’s primary defense mechanism. Manual leukocyte differential counting is labor-intensive, requires skilled microscopy personnel, and is susceptible to human error—particularly when a single specialist must process large volumes of blood smears. This work presents an automated classification algorithm based on K-means clustering, aimed specifically at distinguishing agranulocytes into lymphocytes and monocytes, which are key indicators in the diagnosis of infections such as mononucleosis, tuberculosis, and hepatitis. A central feature of this approach is its emphasis on local technological development, minimizing reliance on costly foreign commercial solutions. Preliminary results confirm the method’s feasibility, achieving an overall accuracy of 77%, with a precision of 93% and a recall of 77%. These outcomes support the advancement of local expertise for transparent, customizable, and accessible biomedical applications.

Natural disasters continue to pose significant threats to human life, often leaving victims trapped under collapsed structures where time-sensitive response is critical. Search and Rescue (SAR) robots have emerged as essential tools to support emergency teams by accessing hazardous or unreachable areas. This paper presents the design, construction, and functional validation of BMO 2.0, a ground-based mobile rescue robot designed to assist in victim localization and environmental monitoring in disaster scenarios. The robot features a modular 3D-printed chassis, optimized for low center of gravity and structural robustness. It integrates a hybrid sensory system, including a thermal camera (MLX90640), a vision module (ESP32-CAM), gas and environmental sensors (MQ-2, aht21), a PIR motion sensor, and an inertial measurement unit (MPU-6050). The core of the system is managed by an Arduino Mega microcontroller and an ESP8266 module for Wi-Fi-based communication with a custom mobile application. The robot was tested in simulated post-disaster environments, demonstrating stable locomotion, real-time wireless data transmission, and accurate thermal and presence detection. The results confirm the viability of BMO 2.0 as a low-cost, adaptable solution for preliminary SAR operations, particularly in regions with limited access to commercial robotic systems.

The integration of trainable molecular algorithms within biological computing frameworks using DNA represents a transformative advance in unconventional computing paradigms. By leveraging the inherent parallelism, high storage density, and biochemical specificity of DNA molecules, these systems enable the execution of computational processes that transcend traditional silicon-based architectures. This work explores recent advances in the design and implementation of DNA-based learning systems capable of performing logical operations, adaptive decision-making, and basic machine learning tasks in biochemical environments. Additionally, the main challenges in the field are addressed, including kinetic unpredictability, molecular noise propagation, limited scalability, and constraints in the interface between molecular and electronic systems. Finally, future perspectives on the training of molecular algorithms are discussed, such as programmable biochemical networks, adaptive molecular memory, and the development of biochemical compilation toolchains. These advances pave the way toward autonomous and intelligent molecular systems with applications in biosensing, therapeutic diagnostics, and next-generation hybrid computing architectures.

This paper presents a dynamic analysis of the ALICE exoskeleton using the Lagrange-Euler formulation. The study begins with a detailed description of the mechanism, highlighting its structural characteristics and mechanical properties. Then, the equations of motion are derived, and the Lagrange-Euler method is applied to model the system dynamics. To validate the proposed model, gait data were collected from five healthy subjects using a custom-developed gait acquisition system. The experimentally obtained data were then used to verify the accuracy and applicability of the derived equations, demonstrating the effectiveness of the proposed approach for analyzing and modeling lower-limb exoskeleton dynamics.

This systematic review analyzes the use of eye-tracking technology for the detection and monitoring of neurodegenerative diseases using PRISMA methodology. Based on a vigorous analysis of 16 high quality studies, the review highlights the potential of eye tracking as a non-invasive diagnostic tool, particularly for diseases such as Alzheimer’s and Parkinson’s disease. The findings indicate high diagnostic accuracy in distinguishing between patients and healthy controls, with significant promise for early detection. Technological advances, including wearable eye trackers and the integration of artificial intelligence with machine learning algorithms such as Random Forest, SVM and CNN, substantially improve the feasibility of its clinical application. However, critical needs for protocol standardization and longitudinal validation are identified to facilitate its widespread adoption in clinical practice.

Equitable access to energy remains a persistent challenge in El Salvador, where despite renewable sources accounting for nearly 70% of the electricity matrix, high levels of energy poverty continue to affect predominantly rural and low-income communities. This paper presents the case of the AD ASTRA 2024 Project, a community intervention that implemented solar lighting and pumping systems in six communities within the metropolitan area of San Salvador, combining technical solutions with training processes and citizen participation. A methodology based on Design Thinking, participatory diagnosis, and sustainable project management was applied, involving the active engagement of youth previously trained in renewable energies. Results demonstrate a significant improvement in perceived security (100%), strengthening of social fabric, and an estimated annual reduction of 5.8 tons of CO2 emissions, with an economic payback period between 3 and 5 years. The experience evidences that solar energy, integrated with community empowerment processes, can act as a catalyst for social, economic, and environmental development in vulnerable contexts, offering a replicable model for the Latin American region.

This paper presents the design and construction of a SCARA (Selective Compliance Assembly Robot Arm) robot aimed at automated assembly tasks. The work addresses mechanical design, actuator selection, control architecture, and experimental validation of the system. The developed prototype enables precise and repetitive operations, making it viable for low-cost production lines. The final result demonstrates that, through the use of accessible technologies such as 3D printing and open-source platforms, it is possible to develop functional prototypes with a high degree of customization and applicability in real manufacturing environments.

This paper presents the development and evaluation of an innovative educational application using Augmented Reality (AR) and robotic kit called “Edarr” that leverages to enhance STEM (Science, Technology, Engineering, and Mathematics) learning among children and young people. The application is designed to introduce foundational concepts in robotics and related STEM areas through interactive and immersive AR experiences. By combining visual engagement with hands-on activities, the project aims to foster curiosity, improve understanding, and face one of the biggest challenges in technology education: bridge the gap between theoretical knowledge and practical application. Initial testing with target age groups indicates increased motivation, higher retention of concepts, and a more inclusive learning environment. This project highlights the potential of AR as a transformative tool in education, particularly for preparing the next generation for the challenges of a technology-driven world.

The digital transformation of the power sector is reshaping the operation, supervision, and planning of distribution networks. In this context, Smart Transformers emerge as a key technology to achieve greater efficiency, reliability, and flexibility in increasingly dynamic and decentralized systems. By integrating real-time monitoring, control, and communication technologies, these devices enable more responsive and intelligent grid management. In Latin America, the adoption of Smart Transformers represents a strategic opportunity to modernize aging infrastructure, traditionally based on conventional transformers with limited visibility and control. However, their implementation faces significant challenges, including high initial costs, outdated regulations, and technical skill gaps. This paper provides a comprehensive overview of Smart Transformers, highlighting their functionalities, key features, and differences compared to traditional transformers based on a literature review and analysis of regional pilot projects. It also analyzes the opportunities and barriers for their deployment in Latin America, drawing on real-world case studies from Brazil, Mexico, and Colombia. Finally, it outlines future perspectives and strategic actions to facilitate a gradual transition toward a more digital, resilient, and interoperable electrical grid.

This project addresses the limitations of current soft gripper end effectors that do not rely on compressed air and struggle to handle fragile foods and solid objects. The lack of research on “soft grippers” based on shape and materials hinders their industrial use and prevents cost savings from eliminating compressed air. To overcome this, a two-finger gripper with a TPU interior will be developed, using the V-Model methodology for structured development and validation. Initial prototypes will be created with PLA, followed by TPU for behavior analysis, with extrusion and line width adjustments to enhance strength and adaptability. A textured surface will be applied for optimal non-slip grip. The results demonstrate that the structure can withstand prolonged use with minimal wear and deformation, even when handling fragile or weighted objects. The project concludes that leveraging 3D printing and the honeycomb principle for soft gripper prototypes significantly improves industrial processing capabilities, offering a cost-effective and efficient solution with potential for future enhancements and diverse applications.

Fused Deposition Modeling (FDM) 3D printing with Thermoplastic Polyurethane (TPU) has recently been used to fabricate soft robotic actuators and grippers, offering an alternative to traditional silicone casting, which is time-consuming and involves complex manufacturing steps. However, 3D printing soft robotic actuators and grippers using highly-flexible TPU (Shore hardness of 60A and 70A) remains unexplored. Although 60A and 70A TPUs are expected to provide excellent features for soft robots, designing and simulating soft robots using these TPU grades requires information about their best-fitting hyperelastic model, parameters, and material properties, which are hitherto unknown. Therefore, we characterize the 60A and 70A TPU behavior using uniaxial tensile tests and identify the best-fitting hyperelastic model with its appropriate parameters. We then demonstrated the feasibility of using these materials by 3D printing high-fidelity soft grippers using 60A and 70A TPU. Compared to a traditional silicone-based soft gripper of a similar size and shape, the proposed 3D printed TPU soft gripper can grasp objects over five times heavier and achieve more than twice the bending angle, while significantly reducing fabrication time and complexity. We also compared the bending angles of 60A and 70A TPU with 85A TPU soft fingers to demonstrate the importance of low-Shore-hardness materials. Furthermore, we compared the experimentally measured bending behavior of the TPU soft fingers with simulation results using the obtained hyperelastic parameters and found that they closely match the experimental results.

Abstract: This systematic literature review examines the state of the art in responsible artificial intelligence (AI) applications within healthcare. The study analyzes the most relevant publications on explainable AI (XAI) implementations and ethical considerations in medicine from 2022 to 2025, providing a comprehensive update to the current body of knowledge. Through screening and selection processes, 21 papers were identified and examined in detail to assess advancements, challenges, and emerging trends in explainable AI adoption across various medical domains including medical imaging, cancer diagnosis and digital pathology. The review highlights explainability methods and implementation barriers for trustworthy AI integration in clinical practice.

Keywords: Artificial intelligence, explainable AI, AI governance, ethical AI, responsible AI, algorithmic ethics, healthcare, medicine, medical care, systematic literature review.

Abstract—The reliable and cost-effective decarbonization of power systems requires innovative strategies to address the
increasing challenge of renewable energy curtailment. This study investigates the application of Autoregressive with Exogenous Variables (ARX) models to predict renewable curtailment under the extreme conditions of the COVID-19 pandemic in Honduras. Usiung daily reports of renewable generation limitations, the data was processed to generate monthly summaries of curtailed energy and identify the most affected plants. The ARX model was simulated with varying polynomial orders to optimize computational efficiency, while Root Mean Square Error (RMSE) and Fit Index (FIT) were employed as evaluation criteria. Results show that the proposed ARX model achieves high accuracy, with prediction errors below 3%, effectively capturing curtailment dynamics at both system and plant levels. Critical network points were identified at the CINCO ESTRELLAS and MARCOVIA plants, underscoring the need to target interconnections to enhance renewable integration. These findings validate the ARX model as a robust forecasting tool for renewable curtailment and highlight its potential to support operational planning, improve system flexibility, and reduce clean energy losses.

Keywords—renewable energy curtailment, ARX model, forecasting, COVID-19, Honduras case study

Existing direct and inverse kinematic models of planar parallel robots assume that the robot’s active joints are all at the bases. However, this approach becomes excessively complex when modeling a planar parallel robot in which the active joints are within one single kinematic chain. To address this problem, our article unveils an alternative for a 3RRR symmetric planar robot modeling technique for the derivation of the robot workspace and the analysis of its direct and inverse kinematics. The workspace was defined using a system of inequalities, and the direct and inverse kinematics models were generated using vectorial analysis and an optimized geometrical approach, respectively. The resulting models are systematically presented and validated. Two final model renditions are delivered supplying a thorough equation analysis and an applicability discussion based on the importance of the robot’s mobile platform orientation. The advantages of this model are discussed in comparison to the traditional modeling approach: whereas conventional techniques require the solution of complex eighth-degree polynomials for the analysis of the active joint configuration of these robots, these models provide an efficient back-of-the-envelope analysis approach that requires the solution of a simple second-degree polynomial.

This study presents a Systematic Literature Review (SLR) on the management of Coffee Berry Borer (CBB) and Coffee Leaf Rust (CLR) in Coffee Agroforestry Systems (CAFS). Despite extensive research on pest and disease management in CAFS, there is a notable absence of literature reviews or bibliometric analyses regarding the study of the CBB and CLR. The novelty of this work lies in offering a comprehensive synthesis of recent research and identifying the technologies employed in these studies, providing valuable insights for future developments in the field. Using the PRISMA protocol and Bibliometrix tool, the research analyzed 51 relevant studies from Web of Science and Scopus databases, covering the period from 2002 to August 2024. The analysis revealed an increasing trend in publications on CBB and CLR in CAFS in recent years. Traditional techniques, such as visual plant observation, pheromone traps, and spore sedimentation traps, remain prevalent. However, the use of advanced technologies in CAFS is still limited. Findings highlight the significant impact of shade on pest dynamics, the importance of biodiversity in pest control, and the need for integrated pest management strategies. The study concludes that while traditional techniques play a crucial role in pest and disease monitoring in CAFS, there is significant untapped potential in integrating advanced technologies such as sensors, UAVs, IoT, robotics, and AI. The limited use of these technologies in CAFS underscores the need for future research to drive their adoption and optimization, aiming to improve sustainable pest and disease management in these complex systems.

This work presents the design and implementation of a data-driven Nonlinear Model Predictive Control (NMPC) framework for an Uncrewed Aerial Vehicle (UAV) equipped with a 3-DOF robotic arm. Real-world data was collected using the Matrice 100 platform and Dynamixel MX-28AR actuators to identify a high-dimensional linear model via Dynamic Mode Decomposition with Control (DMDc), capturing the interactions between the aerial vehicle and the manipulator across 21 state variables. This DMDc-based model is embedded within the NMPC formulation to predict system behavior over finite horizons. The UAV’s orientation is represented using quaternions, enabling continuous and singularity-free attitude control. Additionally, the redundancy of the UAV-manipulator system allows for the integration of secondary objectives into the cost function, supporting flexible task execution. To meet real-time requirements, the control problem is solved using the Acados solver. The resulting controller achieves high-precision tracking while managing internal constraints, demonstrating the potential of data-driven NMPC in aerial manipulation tasks.

In this study, we propose a novel method that integrates Nonlinear Model Predictive Contour Control (NMPCC) with an Exponentially Stabilizing Control Lyapunov Function (ES-CLF) and Exponential Higher-Order Control Barrier Functions to achieve stable path-following and obstacle avoidance in UAV systems. This framework enables uncrewed aerial vehicles (UAVs) to safely navigate around both static and dynamic obstacles while strictly adhering to desired paths. The quaternion-based formulation ensures precise orientation and attitude control, while a robust optimization solver enforces the constraints imposed by the Control Lyapunov Function (CLF) and Control Barrier Functions (CBF), ensuring reliable real-time performance. The proposed method was experimentally validated using a DJI Matrice 100 quadrotor platform, considering scenarios with prior knowledge of obstacle locations. Results demonstrate the controller’s effectiveness in minimizing orthogonal and tangential tracking errors, ensuring stability and safety in complex environments.