DC4: Probabilistic, data-driven sensor fusion for reliable human-robot sensing under real-life conditions

Project: Probabilistic, data-driven sensor fusion for reliable human-robot sensing under real-life conditions (WP2)

Host institution: RWTH (Germany)

Supervisor: Prof. H. Vallery (RWTH)

Co-supervisor(s): Assoc Prof. M. Kok (TUD), Assoc. Prof. Dr. H.J.G. van den Berg-Emons, (EMC)

Objectives:

Create probabilistic sensor fusion algorithms for accurate human-robot motion estimation, integrating all available sensor data and providing uncertainty insights, especially in scenarios where obtaining complete physics-based models is challenging.

Expected Results:

Novel algorithms developed for accurate motion estimates in real-life, non-ideal conditions, addressing challenges such as uncertainty of the sensor locations on the body and fewer sensors than body segments.

Planned secondment(s):

1. TUD (3 months, M17-M19): Collaborate with DC3 to implement sensor fusion, combining data-driven and physics-driven modeling, with M. Kok (KPI: joint conference paper)
2. EMC (3 months, M37-M39): Validation of the motion estimates for a specific use case in the field of rehabilitation medicine with R. van den Berg-Emons and H. Bussmann (KPI: joint journal paper)

Enrolment in Doctoral degree: RWTH Aachen University (Germany)

Required profile: Completed university degree in mechanical engineering, electrical engineering, automation engineering or computer science (master’s degree or comparable).

Desirable skills/interests: Very good knowledge of control and automation engineering or data science including but not limited to fundamentals of sensor fusion and data-driven modelling (ML). Good understanding of mechanics. Previous experience with the integration of IMU sensors and the interpretation of related data. Interest in the crossways between automation engineering and health technologies. Ability to work independently in a scientific context, while exceling in inter-institutional and interdisciplinary collaborations. Good oral and written English skills.