AGAntonio Guerrero-González

Antonio Guerrero-González

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Antonio Guerrero-González has a total of 61 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Automatic Estimation of Manual Sorting Efficiency in Non-Ferrous Scrap Recycling Using Dual-Camera Computer Vision and Deep Learning

    by Antonio Guerrero-González, Tomás Inglés-Muñoz, Francisco García-Córdova, Francisco Antonio Lloret-Abrisqueta

    Manual picking is a critical refining stage in non-ferrous scrap recycling, yet it is rarely measured continuously. This study presents a dual-camera computer-vision system that estimates manual sorting efficiency by counting materials before and after picking. YOLO detectors trained on an in-plant dataset (4361 images, four classes) are combined with a Kalman-filter tracker and line-crossing counters to yield per-class and hourly efficiency indicators. YOLOv11m reached a higher maximum F1-score (0.82 vs. 0.79), and selected models attained mAP@50 values of 0.85–0.86. Image-level splitting may make detector metrics optimistic; stage comparisons are indicative only. In a six-hour test, 3753 input objects and 690 output objects were counted, yielding a count-derived global efficiency estimate of 81.61% (95% Wilson interval, sampling only: 80.34–82.82%). Estimated efficiency exceeded 90% for circuit boards and rubber but was only 75.95% and 73.75% for copper and brass, the most valuable fractions. This demonstrates end-to-end operation, not validated accuracy: no manual reference counts or real output-camera test images were available, so all efficiency values are conditional on the automatic counts. Sensitivity analysis quantifies error propagation into the indicator. The results show the feasibility of detection, tracking, and input–output comparison as a monitoring prototype for Industry 4.0 recycling lines.

  • Open Access

    Local LLMs for Industrial Supervision and Control: An Edge AI Event-Driven Architecture for Proactive Operational Context Management in Real Industrial Environments

    by Fernando Hidalgo-Castelo, Antonio Guerrero-González, Francisco García-Córdova, Francisco Lloret-Abrisqueta, Antonio Piñera-Marín

    Access to operational information in industrial plants forces operators to interrupt their tasks, walk to the human–machine interface (HMI) terminals, and navigate heterogeneous platforms—namely programmable logic controllers (PLC), supervisory control and data acquisition (SCADA) systems, manufacturing execution systems (MES), and enterprise resource planning (ERP) systems—consuming 15–30 min per query. Previous work integrated local large language models (LLMs) into a five-layer cognitive architecture deployed in a precast concrete plant, reducing that time to 14–23 s through voice-based conversational queries; however, model inference accounted for 55.3% of total latency and the system remained reactive. This work incorporates the event-driven paradigm as a non-intrusive augmentation layer that keeps the operational context permanently updated, continuously monitoring the process and refreshing knowledge only when significant changes occur. The architecture is fully local, cloud-independent, graphics processing unit (GPU)-free, and containerized via Docker Compose. Experimental results demonstrate a 26–31% reduction in response times (means of 9.84 s, 11.23 s, and 16.47 s for simple, moderate, and complex queries), an 8.4 °C reduction in peak hardware temperature (from 79.6 °C to 71.2 °C), a 41.6% decrease in thermal variability, and an expansion of the safety margin before central processing unit (CPU) throttling from 5.4 °C to 13.8 °C. The system achieved 100% success rate and availability over 30 min of autonomous operation, validated in a real industrial environment.


  • Open Access

    Artificial Intelligence for Autonomous Vehicles: Robustness Analysis in Complex Urban Traffic Scenarios

    by Brandon Quezada-Godoy, Antonio Guerrero-González, Francisco García-Córdova, Francisco Lloret-Abrisqueta, Antonio Jesús Martínez-Espinosa

    Autonomous driving in complex urban environments remains challenging due to perception uncertainty, dynamic multi-agent interactions, and control instability under adverse conditions. Despite advances in individual components, systematic evaluations of fully integrated modular pipelines under compounded urban disturbances remain scarce. This work presents a modular autonomous driving framework in CARLA Town10HD, integrating Convolutional Neural Network (CNN)-based perception using ResNet-18, global path planning via A* algorithm, and two control strategies: a classical Proportional–Integral–Derivative (PID) controller and a Deep Q-Network (DQN) agent with adaptive geometric steering assistance. A structured protocol assessed robustness across five scenarios: Heavy Rain, Dense Fog, Nighttime Driving, Dense Traffic, and Combined Extreme Conditions. The perception module achieved F1-scores close to 0.99 for traffic-sign, pedestrian, and lane classification; results reflect synthetic CARLA data and should not be interpreted as real-world generalization. The PID controller produced smoother trajectories with lower steering oscillations, while the DQN agent achieved faster traversal times at the cost of higher control variability. Route efficiency remained around 0.96 under isolated disturbances and decreased to 0.52 under compounded conditions, confirming sensitivity to multi-factor complexity. This study contributes a reproducible multi-scenario benchmark quantifying stability–adaptability trade-offs between classical and learning-based control, identifying scenario generalization and simulation-to-reality transfer as key future directions.

  • Open Access

    FPGA-Based CNN Acceleration on Zynq-7020 for Embedded Ship Recognition in Unmanned Surface Vehicles

    by Abdelilah Haijoub, Aissam Bekkari, Anas Hatim, Mounir Arioua, Mohamed Nabil Srifi, Antonio Guerrero-Gonzalez

    Unmanned surface vehicles (USVs) increasingly rely on vision-based perception for safe navigation and maritime surveillance, while onboard computing is constrained by strict size, weight, and power (SWaP) budgets. Although deep convolutional neural networks (CNNs) offer strong recognition performance, their computational and memory requirements pose significant challenges for deployment on low-cost embedded platforms. This paper presents a hardware–software co-design architecture and deployment study for CNN acceleration on a heterogeneous ARM–FPGA system, targeting energy-efficient near-sensor processing for embedded maritime applications. The proposed approach exploits a fully streaming hardware architecture in the FPGA fabric, based on line-buffered convolutions and AXI-Stream dataflow, while the ARM processing system is responsible for lightweight configuration, scheduling, and data movement. The architecture was evaluated using representative CNN models trained on a maritime ship dataset. Our experimental results on a Zynq-7020 system-on-chip demonstrate that the proposed co-design strategy achieves a balanced trade-off between throughput, resource utilisation, and power consumption under tight embedded constraints, highlighting its suitability as a practical building block for onboard perception in USVs.


  • Open Access

    YOLOv5: an improved algorithm for real-time detection of industrial defective pieces

    by Abdelaziz Elbaghdadi, Yassine Yazid, Ahmed El Oualkadi, Antonio Guerrero-González, Soufiane Mezroui

    The rapid advancement of communication technologies and the growing demand for artificial intelligence are transforming traditional manufacturing into smart industries. Robotic arms and smart vision cameras are widely adopted to support industrial internet of things (IIoT) applications. Beyond enhancing production efficiency and quality, these technologies play a crucial role in cost reduction, energy savings, and improving operator safety. In this article, we propose an intelligent industrial system using an improved version of the you only look once (YOLO) algorithm for defect detection on production lines. The system integrates robots and cameras to automate defect inspection and classification of manufactured pieces. An updated YOLOv5 model is designed as an end-to-end solution for detecting surface defects in three specific regions. We trained and evaluated the model using custom data tailored to the inspected pieces. The system achieved a 99% mean average precision (mAP) and an 80% recall rate. Additionally, it delivers a 99% detection rate at high speed, enabling real-time surface defect detection. This method not only accurately predicts defective locations but also provides size information, which is critical for assessing the quality of newly produced pieces.

  • Open Access

    Multimodal Cognitive Architecture with Local Generative AI for Industrial Control of Concrete Plants on Edge Devices

    by Fernando Hidalgo-Castelo, Antonio Guerrero-González, Francisco García-Córdova, Francisco Lloret-Abrisqueta, Carlos Torregrosa Bonet

    Accessing operational information across industrial systems (ERP, MES, SCADA, PLC) in concrete plants requires 15–30 min and specialized knowledge. This work addresses this accessibility gap by developing a conversational AI system that democratizes industrial information access through natural language. A five-layer cognitive architecture was implemented integrating the Mistral-7B model quantized in GGUF Q4_0 format (3.82 GB) on a Raspberry Pi 5, Spanish speech recognition/synthesis, and heterogeneous industrial protocols (OPC UA, MQTT, REST API) across all automation pyramid levels. Experimental validation at Frumecar S.L. (Murcia, Spain) characterized performance, thermal stability, and reliability. Results show response times of 14.19 s (simple queries, SD = 7.56 s), 16.45 s (moderate, SD = 6.40 s), and 23.24 s (complex multilevel, SD = 6.59 s), representing 26–77× improvement over manual methods. The system maintained average temperature of 69.3 °C (peak 79.6 °C), preserving 5.4 °C margin below throttling threshold. Communication latencies averaged 8.93 ms across 10,163 readings (<1% of total latency). During 30 min of autonomous operation, 100% reliability was achieved with 39 successful queries. These findings demonstrate the viability of deploying quantized LLMs on low-cost edge hardware, enabling cognitive democratization of industrial information while ensuring data privacy and cloud independence.


  • Open Access

    Integration of Industry 5.0 Technologies in the Concrete Industry: An Analysis of the Impact of AI-Based Virtual Assistants

    by Carlos Torregrosa Bonet, Francisco Antonio Lloret Abrisqueta, Antonio Guerrero González

    The construction industry, traditionally lagging behind in terms of digitalization, faces significant challenges in its transition to Industry 4.0, which is characterized by the use of advanced technologies such as artificial intelligence (AI), the Industrial Internet of Things (IIoT), and cloud computing. This article presents the development and implementation of an AI-based virtual assistant, designed to optimize the operation and maintenance of concrete production plants. The assistant helps reduce the margin of human error, improve operational efficiency, and facilitate continuous training for operators. These advancements foster a more collaborative and digitalized environment, while also generating environmental, economic, and social benefits: reduced material and energy waste, lower carbon footprint, increased workplace safety, and strengthened organizational resilience. The results show high accuracy in voice transcription (96%) and a 100% success rate in responding to technical queries, demonstrating its effectiveness as a support tool in industrial settings. Based on these findings, it is concluded that the incorporation of AI-based virtual assistants promotes a more sustainable and responsible production model, aligned with the Sustainable Development Goals of the 2030 Agenda, and anticipates the principles of Industry 5.0 by promoting symbiotic collaboration between humans and technology. This innovation represents a key advancement in transforming the concrete industry, contributing to productivity, environmental sustainability, and workplace well-being in the sector.


  • Redefining Human-Machine Collaboration: Industry 5.0 to Improve Safety and Efficiency

    by Francisco Antonio Lloret Abrisqueta, Antonio Guerrero González, Roberto Zapata Martinez

    This study presents an innovative implementation of Industry 5.0 principles in a window production line, integrating advanced robotics and artificial intelligence technologies to improve operational efficiency and worker well-being. A robotic cell was designed to automate the handling of heavy components in the final production stage, resulting in a 35% reduction in cycle times and a significant decrease in ergonomic risks. Additionally, an interactive voice assistant based on generative AI was implemented, allowing operators to access system data and technical information in real-time through cognitive interaction. The results show a substantial improvement in job satisfaction, with a 278% increase in the perception of occupational health. This approach not only optimizes productivity but also redefines workers' roles, aligning with the human-centered vision of Industry 5.0. The study demonstrates how the integration of advanced technologies can create safer, more efficient, and adaptable work environments in modern manufacturing.


  • Open Access

    Enhanced YOLOv8 Ship Detection Empower Unmanned Surface Vehicles for Advanced Maritime Surveillance

    by Abdelilah Haijoub, Anas Hatim, Antonio Guerrero-Gonzalez, Mounir Arioua, Khalid Chougdali

    The evolution of maritime surveillance is significantly marked by the incorporation of Artificial Intelligence and machine learning into Unmanned Surface Vehicles (USVs). This paper presents an AI approach for detecting and tracking unmanned surface vehicles, specifically leveraging an enhanced version of YOLOv8, fine-tuned for maritime surveillance needs. Deployed on the NVIDIA Jetson TX2 platform, the system features an innovative architecture and perception module optimized for real-time operations and energy efficiency. Demonstrating superior detection accuracy with a mean Average Precision (mAP) of 0.99 and achieving an operational speed of 17.99 FPS, all while maintaining energy consumption at just 5.61 joules. The remarkable balance between accuracy, processing speed, and energy efficiency underscores the potential of this system to significantly advance maritime safety, security, and environmental monitoring.


  • Open Access

    Interactive Virtual Reality Application for 3D Disassembly and Assembly: Innovation in Industrial Technical Education

    by Mario Cárdenas Bastán, Antonio Guerrero González, Francisco Lloret Abrisqueta

    This study presents the development of a cutting-edge virtual reality (VR) application aimed at education in industrial and technical environments. Using the Meta Quest 3 system and the ShapesXR platform, the application allows users to interactively disassemble and assemble 3D models in an immersive environment. Detailed models were developed through parametric modeling tools such as Autodesk Fusion 360 and 3D optimization software like Blender, optimized for a high-performance VR environment. The feasibility of the platform as an innovative educational tool was evaluated, and areas for improvement for future iterations of the system were identified.