Héctor Azpúrua
Assistant Professor, Department of Computer Science (DCC), Universidade Federal de Minas Gerais (UFMG)
Co-founder, MIRA — Mapping and Robotics
Belo Horizonte, Brazil
I am an Assistant Professor in the Department of Computer Science (DCC) at the Universidade Federal de Minas Gerais (UFMG), Brazil. I work in the Computer Vision and Robotics Laboratory (VeRLab) and take part in GEAR, the department’s advanced robotics study group. I am also a co-founder of MIRA, a Brazilian company that builds LiDAR survey tools, 3D mapping systems and inspection robots for caves, ducts and dam galleries.
My research is about robots that go where people should not: autonomous exploration, mapping and navigation in confined, subterranean and industrial environments, with ground, aerial and multi-robot systems.
I received my PhD in Computer Science from UFMG in 2022. My thesis on terrain-aware autonomous exploration won the CAPES Thesis Award for the best PhD thesis in Computer Science in Brazil (2023). Before joining the faculty in 2024, I was a Machine Learning Engineer at Meta, a computer scientist at Komatsu, and a researcher at the Instituto Tecnológico Vale (ITV), where I coordinated work on the EspeleoRobô, a robot for inspecting caves and confined spaces.
Research
- Autonomous exploration and mapping of confined and subterranean spaces: caves, mines, galleries and ducts.
- Localization, SLAM and navigation for field and service robots, including low-cost sensing for industrial plants.
- Multi-robot, swarm and aerial systems, from communication-aware planning to search and rescue.
- Robot learning: reinforcement learning and world models for navigation, manipulation and humanoids.
Prospective students
I am looking for MSc, PhD and undergraduate research (IC) students interested in robotics. If you would like to work with me, e-mail hector.azpurua@dcc.ufmg.br with your CV and a few lines on the problems you want to work on. See the group page for current projects.
News
| 2026 | Four papers at CROS 2026: adaptive swarm navigation, a fiber-optic tether for robots in confined spaces, UAV person detection over Wi-Fi HaLow, and reward shaping for a robot manipulator. |
|---|---|
| 2026 | Serving on the program committee of the Latin American Robotics Symposium (LARS 2026) and on the organizing committee of CROS 2026. |
| 2026 | New in the Journal of Intelligent & Robotic Systems: terrain classification for ground robots with an instrumented wheel. |
| 2025 | Co-organized the 1st Brazilian Conference on Robotics (CROS 2025), held at UFMG. |
| Sep 2024 | Joined the Department of Computer Science at UFMG as an Assistant Professor. |
| 2024 | Received the Most Influential Paper Award of the Journal of Control, Automation and Electrical Systems (Brazilian Society of Automation). |
| 2023 | My PhD thesis on terrain-aware autonomous exploration received the CAPES Thesis Award for the best PhD thesis in Computer Science in Brazil. |
| 2022 | My PhD thesis took 1st place in the Robotics Thesis Contest at IEEE LARS 2022. |
Selected publications [full list]
- RASA Survey on the autonomous exploration of confined subterranean spaces: Perspectives from real-word and industrial robotic deploymentsRobotics and Autonomous Systems 2023
- AURO
- T-ASEEKF-LOAM: An Adaptive Fusion of LiDAR SLAM With Wheel Odometry and Inertial Data for Confined Spaces With Few Geometric FeaturesIEEE Transactions on Automation Science and Engineering 2022
- JINTTowards Semi-autonomous Robotic Inspection and Mapping in Confined Spaces with the EspeleoRobôJournal of Intelligent & Robotic Systems 2021
- JFRCooperative digital magnetic‐elevation maps by small autonomous aerial robotsJournal of Field Robotics 2019
- ICRAThree-dimensional Terrain Aware Autonomous Exploration for Subterranean and Confined SpacesIn IEEE International Conference on Robotics and Automation (ICRA 2021)