Our Mission
Our mission is to contribute to the advancement of robotics through world-class scientific research and educating the next generation of roboticists capable of developing innovative technologies that address real-world challenges.
We believe that robotics is a strategic area for Argentina’s scientific, technological, and productive development. Therefore, we investigate the latest trends in related subjects such as artificial intelligence, perception, computer vision, and autonomous systems, fostering collaboration and exchange with the local and international scientific community.
At the same time, we are committed to ensure that the results of our research extend beyond academia, working on innovation and the transfer of knowledge and technology to industry and society.
Our History
The Laboratory of Robotics and Embedded Systems (LRSE) was founded in 2009 within the Department of Computer Science at the Faculty of Exact and Natural Sciences, University of Buenos Aires, with the goal of rebuilding and strengthening robotics research and development within the Faculty. Since then, the laboratory has experienced sustained growth, becoming one of Argentina’s leading robotics research groups. Over the years, the LRSE has accomplished high-impact technological projects, established collaborations with other robotics research groups both locally and internationally, and published its scientific contributions in the world’s leading robotics journals and conferences. In parallel, the laboratory has educated more than a dozen PhD roboticists who have earned their PhDs and gone on to successful careers in academia and industry, both in Argentina and around the world.
Our Focus
- Our research focuses on the development of algorithms and methods that enable robots to perceive, understand, and autonomously interact with the real world. We work on advanced Simultaneous Localization and Mapping (SLAM) systems, including visual, visual-inertial, and LiDAR-based approaches for both indoor and outdoor environments, using 2D and 3D representations. We investigate techniques for three-dimensional scene reconstruction, aerial photogrammetry using drones, and computer vision to generate highly accurate digital models, while incorporating semantic information to build richer representations of the environment. We develop autonomous navigation algorithms based on geometric methods, visual appearance, and teach-and-repeat strategies, enabling ground and aerial robots to operate robustly and autonomously in large-scale real-world environments. Finally, we explore the use of modern artificial intelligence techniques for robot control, including deep reinforcement learning for the locomotion of legged robots, such as quadrupeds and hexapods, and imitation learning for robotic manipulators capable of acquiring new skills from human demonstrations.
