Verbetering van RSS-gebaseerde optische draadloze positionering via machinaal leren, fysisch modelleren en IMU-integratie

The work will start with a detailed study of the challenges in precise positioning in indoor environments, and the potential in the combination and cooperation of many sensors and techniques. The PhD student will hence develop strategies for precise positioning leveraging on sensor federations and electronics dispersed in the environment. This will be based on theoretical and simulation-based analyses. Next, experiments will be set-up and carried out in the hybrid sensor testbed that is under construction in the DRAMCO lab. These experiments will validate the new positioning solutions. Also, they will deliver essential data for the investigation of machine learning techniques to further improve the performance.