Direkt zum Inhalt
Symbolfoto: Das AIT ist Österreichs größte außeruniversitäre Forschungseinrichtung

AIT: First Place in the SLAM Benchmark at the IROS 2026 Robotics Conference

18.11.2026
How robots navigate when GPS fails: AIT develops a solution for precise 3D navigation and real-time mapping.
1_Mit SLAMLab erzeugte LiDAR-Punktwolke, eingefärbt nach der Intensität des zurückgestreuten Lichts. Copyright AIT

1_Mit SLAMLab erzeugte LiDAR-Punktwolke, eingefärbt nach der Intensität des zurückgestreuten Lichts. Copyright AIT

The AIT Austrian Institute of Technology took first place in the COMFORT Localization Benchmark at the international robotics conference IROS 2026 in Pittsburgh, USA, with its SLAMLab solution. SLAMLab delivered the most accurate 3D localization and mapping of all participating systems and was honored with the Industry Award.

For robots and autonomous machines to move reliably, they must continuously determine their position and orientation. Outdoors, this is usually done via satellite navigation (GNSS) such as GPS. However, inside buildings, tunnels, and underground parking garages, in narrow urban canyons, under bridges, or in forests, the satellite signal is disrupted or completely unavailable. This is exactly where SLAMLab comes in: The system determines its position directly from data collected by its onboard sensors and simultaneously creates a precise 3D map of the environment, even in places where GNSS fails.

First Place in International Competition for AIT’s SLAMLab Solution

At the COMFORT Localization Benchmark held during the international robotics conference IROS 2026 in Pittsburgh, USA, approximately 50 teams from research and industry competed with their SLAM (Simultaneous Localization and Mapping) methods. The evaluation focused on how accurately the systems could determine the movement of the four-legged ANYmal robot while simultaneously creating a 3D map of its surroundings – for example, inside buildings, in smoky conditions, and on rough terrain.

The AIT solution, SLAMLab, achieved the highest accuracy and tops the benchmark rankings.

AIT SLAMLab: Real-Time Precision, Without a GPU

Until now, the highest precision in mapping has usually been achieved only retrospectively through time-consuming data analysis in the office (post-processing). Real-time SLAM systems, on the other hand, often deliver only moderate accuracy or require powerful computing hardware such as GPUs.

AIT SLAMLab goes one step further: It combines real-time data acquisition, high measurement accuracy, and flexible sensor integration into a user-friendly framework. The system combines sensors such as LiDAR (Light Detection and Ranging), IMU (Inertial Measurement Unit), and GNSS (Global Navigation Satellite System) and automatically adapts when individual sensors fail. It is also specifically optimized for resource-constrained, more cost-effective computing hardware.

From Research to Industrial Application

The combination of measurement accuracy, real-time capability, and moderate hardware requirements opens up a wide range of applications, such as the mobile surveying of roads and rail lines or the inspection of technical infrastructure. In practice, SLAMLab is already being used successfully – from navigating an autonomous forklift in the AIT Large-Scale Robotics Lab to an assistance system for collision avoidance for large construction machines, to 3D mapping of buildings.

Outlook: Multi-Agent Collaboration with a Shared Map

SLAMLab is undergoing continuous further development at AIT. A future area of development is collaborative mapping, in which multiple autonomous systems jointly build and use a global 3D map. This approach is intended to enable the efficient mapping of even larger and more complex operational areas.

 

AIT Center for Vision, Automation & Control: https://www.ait.ac.at/en/about-the-ait/center/center-for-vision-automation-control

AIT SLAMLab Fact Sheet: AIT SLAMLab

SLAMLab Website: https://slamlabapp.github.io/latest/

SLAMLab on YouTube: https://www.youtube.com/@SLAMLabApp

About AIT

Mag. Michael Hlava
Head of AIT Communication
T +43 (0)50550-4014 I michael.hlava(at)ait.ac.at

Daniel Pepl, MAS MBA
AIT Communication / AI Factory Austria AI:AT
T +43 (0)50550-4040 I daniel.pepl(at)ait.ac.at

Mag. Florian Hainz BA
AIT Communication - Press & PR
T +43 (0)50550-4518 I florian.hainz(at)ait.ac.at