Conference Agenda

Bipedal locomotion, such as walking or running, involves complex coordination of rhythmic and cyclic movements that must be stable, smooth, and adaptable to varying terrains, which is challenging to achieve in robotic and simulated environments. Current reinforcement learning (RL) approaches often fail to generate stable and natural locomotion patterns due to a lack of inherent rhythmic control, resulting in jerky, unstable, and inefficient gaits. These methods typically do not incorporate the biological principles of rhythmicity and adaptability, which are crucial for achieving natural bipedal locomotion. This study presents a novel approach integrating Central Pattern Generators (CPG) with multiple RL algorithms, including Maximum a Posteriori Policy Optimization, Deep Deterministic Policy Gradient, and Soft Actor-Critic (SAC), using Matsuoka Oscillators to generate rhythmic patterns. By comparing these RL-CPG hybrid methods, the research demonstrates improvements in energy efficiency and synchronization for SAC+CPG in controlled environments. While other algorithms may have advantages in different conditions, SAC+CPG showed the most stable, and rhythmic gait, while minimising energy usage under the tested parameters. This study highlights the first multi-algorithm application of RL combined with CPGs for rhythmic control in bipedal locomotion, contributing to the future of robotics and cyber-physical systems.

This paper presents the design, development, and validation of the DeepBanana platform, an artificial intelligence (AI)-based solution for the early detection of diseases in banana crops through automated analysis of leaf images. Framed within the international DeepFarm project, funded by the Erasmus+ program, the system integrates convolutional neural networks (CNNs), data augmentation techniques, transfer learning, and a modular architecture adaptable to the technological conditions of Dominican farms. The platform was trained on a labeled dataset of over 1,900 images classified into seven plant health categories, achieving an accuracy close to 89%. The technical pipeline stages, CNN model architecture, automated retraining system, and the incorporation of a conversational interface with retrieval-augmented generation (RAG) capabilities are detailed.

Concentrations of particulate matter (PM) in the air pose a significant risk to human health and the environment. Accuracy in the measurement of these pollutants is critical for effective air quality management; however, monitoring stations present errors and inconsistent data that affect the reliability of the analysis. In this study, different methods based on data science and machine learning are presented and compared to correct and improve the quality of PM measurements. This includes an exploration data analysis to identify temporal patterns in air pollution specifically of PM, detection and removal of outliers using the interquartile range method, normalization and transformation of temporal variables, and implementation of a convolutional autoencoder model for missing data correction. The methodology was applied to a dataset collected by a monitoring station in Panama, and the results showed that the removal of outliers significantly reduced the distortion in the data, while the autoencoder achieved a moderate reconstruction of missing values, with a MAE of 0.1322 and a coefficient of determination R² of 0.5770.

The findings suggest that the combination of statistical techniques and machine learning models allows to improve the reliability of PM monitoring data, providing more accurate information for environmental decision-making. In addition, this study opens new lines of research, such as the development of low-cost correction models for community stations, the analysis of the impact of meteorological events on particulate matter concentrations, and the comparison of pollution patterns in different urban environments.

Effective attendance management and registration systems are essential for the successful organization of cultural and institutional events. This study presents the design of a system developed using PHP Laravel and MySQL, leveraging QR code technology for accurate, secure, and real-time attendance tracking. Following the agile Scrum methodology, the system undergoes iterative improvements based on user feedback, while its UX/UI-driven design ensures an intuitive and accessible interface for organizers and attendees. In addition to automating administrative tasks, the system incorporates real-time data analytics to monitor attendance, evaluate participation, and enhance event planning. Secure data storage is ensured through MySQL integration, adhering to protection standards via JWT, OAuth2, and Google authentication. Designed for efficiency, scalability, and usability, the system addresses the needs of the Department of Art and Culture at Don Bosco University, where future implementation is envisioned, ensuring its practical applicability in optimizing resource allocation and event organization across various sectors.

Literature review is crucial for research development, but the growing number of digital publications has complicated this process, increasing the risk of bias. This article aims to develop and apply an automated academic literature review approach using text mining techniques in the R programming language. A database of 86,820 articles published in scientific journals, containing the search terms ("model" AND "growth" AND "economic"), hosted in the open access database Redalyc, was retrieved. A programming syntax was developed that optimized data download, processing and analysis, allowing its replication in future literature review processes. This study demonstrates the potential of text mining tools and automated bibliometric analysis, using R, to optimize literature review in the scientific field.

The rapid integration of generative artificial intelligence (AI) systems and AI-driven tools in social and professional domains has made AI literacy a critical competency for students, educators, researchers, and administrators in higher education. This study synthesizes current research on AI literacy frameworks in the context of universities, highlighting their dimensions, units of analysis, gaps, and emerging trends. A systematic literature review was conducted, focusing on the essential components of an AI literacy framework for higher education. The search equation, developed using a generative AI tool, was applied to the Scopus database and the AI-powered search engine Undermind, yielding 173 documents. Inclusion and exclusion criteria were applied, resulting in 20 articles for analysis. The findings reveal a growing trend towards interdisciplinary and competency-based approaches to foster AI literacy in higher education. Key components identified include technical knowledge, ethical considerations, social responsibility, critical thinking, collaboration, innovation, and adaptability. However, challenges persist, such as the need for empirical validation of conceptual models, adaptation to rapidly evolving AI technologies, and addressing equity gaps. Future research should focus on integrating universal and personalized frameworks, minimizing equity gaps, and iteratively adapting to the implications of AI advancements. This study provides a foundation for exploring strategies to equip higher education stakeholders with the necessary AI literacy skills to navigate and thrive in an AI-driven future.