Competencies
Responsible Researcher: José Santos
The Flexible Manufacturing Systems Laboratory (LSFP) is dedicated to developing advanced solutions for industrial automation, real‑time quality inspection and intelligent systems for electric micromobility. The research combines computer vision, artificial intelligence and embedded electronics to create prototypes ready for integration into production lines and innovative products.
Core Competencies
i) Automated visual inspection using deflectometry, projecting patterns onto surfaces to reveal deformation associated with defects.
ii) Real‑time image acquisition and processing, with deep‑learning‑based detection and classification algorithms (YOLO) implemented in Python/OpenCV.
iii) Automated decision-making (“accept/reject”) based on predefined criteria, with HMI interface and report generation.
iv) Development of intelligent controllers for e‑bikes, integrating pedal, brake and speed sensors with ESP32 microcontrollers and BLDC drivers.
v) Advanced energy management via BMS (monitoring voltage, temperature and current), with algorithms for autonomy and safety optimization.
vi) Predictive maintenance of equipment using machine‑learning techniques.
Application Examples
1) Predictive maintenance of machining centres and cutting tools
The LSFP develops automated systems for inspecting painted components, eliminating ergonomic limitations of manual inspection and reducing human error. The process follows a structured workflow:
- Pattern projection (deflectometry) onto the surface
- Image acquisition using a camera
- Processing with YOLO algorithms for defect detection and classification (scratches, dents, paint lack/excess)
Automatic decision (“OK/NOT OK”) based on predefined criteria
Technologies: Python, OpenCV, convolutional neural networks.
2) Intelligent controller for e‑bikes
The laboratory develops controllers that adjust electrical assistance based on sensor data (pedal, brake, speed) and information from the BMS. The architecture includes an ESP32, a BLDC driver and mobile‑device communication for testing. Integrated algorithms reduce operating temperature, increase battery autonomy and extend battery life, ensuring user safety.
Key technologies:
- Computer vision: deflectometry, YOLO, Python, OpenCV
- Embedded electronics: ESP32, BLDC drivers, BMS integration
- Algorithms for improved Remaining Useful Life (RUL) prediction