Computer Vision aims to introduce students to the study of computer vision and design of digital image processing algorithms. The theoretical concepts explained in the lectures are programmed later in the laboratory in Python with OpenCV.

The subject of Perception and Control introduces students to the design of sensing systems and control techniques applied, as illustrative example, to a mobile robot. Its objectives are the study and design of various types of sensors and actuators, the acquisition of information from these sensors, the introduction to control theory and case studies of perception and control systems.

Multiple neural networks are employed to detect and classify the driver's state and actions in an autonomous system. The challenges of implementing these models on low-power embedded devices are also examined.

The main objective of this work is to develop a comprehensive and robust navigation architecture in ROS 2 for an autonomous race car capable of operating in both time-trial scenarios and races against opponents.

This course introduces students to advanced perception techniques for understanding scenes based on data provided by robot sensors. Both classical and deep learning (DL) approaches using neural networks (NNs) are addressed, primarily from camera data. Theoretical concepts explained are practically validated in the laboratory. The topics covered include: introduction to perception systems, computer vision, NNs for perception, advanced NN architectures and applications, and object tracking. Upon completing the course, students will be able to utilize NNs to solve robotic problems applicable to other relevant engineering domains.