The Department of Engineering Cybernetics at the Norwegian University of Science and Technology (NTNU) invites applications for a PhD position in Resilient Multi-Modal Scene Understanding and Affordance Mapping. This research opportunity is part of the Norwegian Centre for Embodied AI (NCEI), a national initiative uniting leading robotics and AI groups to advance the science of embodied intelligence. The project focuses on developing robust frameworks for scene understanding and mapping in environments where conventional positioning signals are unavailable and perception is intermittently degraded, such as underwater or in challenging above-water settings.

The successful candidate will work on fusing heterogeneous sensing modalities using uncertainty-aware probabilistic optimization, maintaining semantic, structural, and affordance-level understanding of dynamic environments. The research will emphasize adaptive fusion processes based on modality confidence, construction of rich scene graphs encoding geometry, semantics, functional relationships, and actionable affordances, and predictive reasoning to bridge gaps in observation. An optional extension includes incorporating learned world-model components to anticipate future sensory states and provide stabilizing priors. The anticipated outcome is a robust, predictive mapping and understanding system capable of sustaining coherent situational awareness and reasoning about possible robot-environment interactions under partial observability.

Supervision will be provided by Professor Annette Stahl (main supervisor) and Professor Edmund Brekke (co-supervisor). The position is embedded in a vibrant, international research community with access to world-class facilities and strong collaborations across Norwegian universities, research institutes, industry, and public agencies. The NCEI’s mission is to chart a generalizable path for physical AI, transforming how robot morphology and autonomy are co-designed for diverse operational environments and missions.

Eligibility: Applicants must hold a Master’s degree in cybernetics, control systems, or equivalent, with strong training in control of marine vehicles and/or robots. Additional training in machine learning is advantageous. Candidates should have a strong academic record (B or better on the NTNU scale), solid programming skills, and a strong background in mathematics and robot perception. Non-native English speakers or those without a Scandinavian language must provide proof of English proficiency (TOEFL, IELTS, or Cambridge certificates). Final-year master’s students may apply, but employment will be postponed until the degree is completed.

Funding: The position offers a gross annual salary starting from NOK 550,800, with a 2% statutory contribution to the State Pension Fund. The employment period is 3 years, and the position is conditional on external funding. As a PhD candidate at NTNU, you will benefit from career guidance, a mentor program, and access to employee benefits.

Application Process: Applications must be submitted electronically via Jobbnorge.no, including transcripts and diplomas, CV, a short motivation letter, and contact information for at least two referees. All documents must be in Norwegian, a Scandinavian language, or English. Only applications received by the deadline (16 December 2025) will be considered. For further details, visit the application link provided.

NTNU is committed to diversity and encourages applications from all qualified candidates, regardless of gender, background, or career interruptions. The university offers a supportive and inclusive work environment in Trondheim, a city known for its rich cultural scene, excellent quality of life, and outstanding opportunities for education and recreation.