PhD on Multipath-based Cooperative SLAM through Machine Learning

This doctoral position focuses on investigating machine learning techniques to advance simultaneous localization and mapping (SLAM) by harnessing multipath radio propagation across cooperative wireless devices. Based at ENSEA and associated with the LETI laboratory, the candidate will extract propagation channel features from multiple radio links to jointly estimate the positions of mobile devices and nearby scatterers. The work combines methodological development with empirical validation using ultra-wideband real-world measurements and synthetic ray-tracing simulations. The successful applicant will contribute to innovative navigation solutions applicable to complex and dynamic environments such as robotics fleets and emergency response teams.

• Investigate machine learning models to solve cooperative simultaneous localization and mapping problems using radio signals.
• Identify and extract propagation channel features from multiple wireless links, focusing on multipath echo arrival times.
• Develop algorithms that exploit multipath correlation and spatial diversity across moving cooperative devices.
• Evaluate proposed algorithms using real-world measurements collected with ultra-wideband hardware in reference indoor settings.
• Generate and utilize synthetic datasets via ray-tracing simulators for extensive testing and validation.
• Collaborate with research team members on navigation applications for complex environments including drones, mobile robots, and emergency personnel.