Abstract:

Understanding and monitoring all of the world's oceans is a vital endeavour for all life on earth. It is favourable to employ mechatronic systems for performing these tasks such as mapping the seafloor. With increasing interest in the recent years, more and more Autonomous Underwater Vehicles (AUVs) are being used for that requiring sophisticated modelling and control software. Due to the lack of GPS, underwater navigation for AUVs is an ambitious task which is commonly solved as a general state estimation problem using, among others, acoustic and inertial measurements. Assuming high-quality devices combined with compact acoustic positioning sensors, it is examined which navigation accuracy can at best be expected for a typical deep-sea scanning mission with focus on adding a dynamic model from the AUV's low-level control as an additional virtual measurement.

Bio:

David Oertel studied mathematics as well as computer science at KIT and finished his studies with a dual diploma. He recently received his Ph.D. at IAR-IPR and has been researching localization and navigation for UAVs in the group for Micro and Collaborative Robotics, but has started to shift his focus to industrial applications recently. He is currently also studying electrical enginieering with the goal of getting a B.Sc.E.E.