PhD in Visual SLAM for construction in Luxembourg

The position holder will work under the umbrella of a partnership project Theia with the Luxembourg-based company GAMMA AR, developing a solution for real-time vision-based Simultaneous Localization and Mapping (SLAM) of a hand-held tablet on real construction sites (i.e. as-built), with drift-free alignment with existing Building Information Management (BIM) data (i.e. as-planned). This solution will extend novel Situational Graphs (S-Graphs), concretely Informed S-Graphs and Vision-based S-Graphs. The main scientific challenges are the (1) vision-based global localization of the hand-held device within BIM (i.e. global alignment between the real construction site and the BIM) when the main structural elements (e.g. walls) are built and there are no major deviations from the BIM. (2) vision-based individual alignment between the main elements of the construction site and BIM, when there are no major deviations. And (3) vision-based global and individual alignment between the construction site and its main elements and BIM in the presence of deviations. The solution will be validated experimentally in real datasets and real construction sites (together with their BIM). On top of this, the solution will be integrated into ready-to-use robotic platforms to demonstrate its potential use in autonomous robot systems in real-world use cases.

Duties of the position

The appointee will be expected to conduct research on perception and situation understanding, making contributions to the state-of-the-art in the fields of simultaneous localization and mapping (SLAM), computer vision, machine learning, robotics, and/or related; propose and implement working novel solutions validated in real datasets and real construction sites and our robotic platforms; contribute to the development of the solution and its technological demonstration envisaged on the partnership project Theia with Gamma AR, its success and impact; as well as the research line of situational awareness for robots and autonomous systems of the ARG; and disseminate results through scientific publications in top peer-reviewed international journals, such as IEEE RA-L, and Elsevier RAS, and conferences, such as ICRA, and IROS.

Required selection criteria

Applicants should possess a master’s degree in Computer Science, Robotics, Electrical Engineering, Applied Mathematics, Artificial Intelligence, Mechanical Engineering, or a related field; background in (at least one) simultaneous localization and mapping (SLAM), computer vision, machine learning, robotics, and/or related; strong mathematical background; strong programming skills in (at least one) Python/C++; familiarity with computer vision frameworks like OpenCV, and/or deep learning frameworks such as PyTorch and TensorFlow, and/or optimization-based frameworks such as GTSAM, G2O, or similar, and/or robotics frameworks like ROS; fluent written and verbal communication skills in English; commitment, high motivation, and high ambition; team working abilities and critical mind. Experience in Android programming is a plus.

Benefits 

The selected candidate will be offered a 12 month-salary package, over six weeks of paid time off, meal vouchers and health insurance; free language classes; career resources and extracurricular activities; support to acquire funding from national, European and private sources; exciting infrastructures and unique labs and more.

Organisation/Company – University of Luxembourg

Research Field –

Researcher Profile –

Country – Luxembourg

Application Deadline – 2 June 2025