Colloquium "Vision aided inertial navigation using single video camera measurements"

Abstract:

Future unmanned aerial vehicles have a need of high levels of autonomy and independence. They are going to become small, very agile and highly maneuverable. These requirements are critically connected to the sensing, computational and interpretational competence of the vehicles. Integrated navigation system design requires selection of a set of sensors and computation power that provides reliable and accurate navigation parameters (position, velocity and attitude) with high update rates and bandwidth in small and cost effective manner. Inertial and vision sensors with their complementary characteristics have the potential to meet these requirements. The information for updating the navigation parameters (position, velocity and attitude) from the inertial sensors comes from the act of combining vision with motion (manoeuvres). Using such a combination gets around many of the cons of the single video camera. Manoeuvres appear to be essential for certain or desired accuracy of the navigation parameters. Simply passing or flying by a feature (map point) with no manoeuvre will not help much for autonomous navigation. These facts bring new challenges to the practical design of today’s modern jam proof GPS free autonomous navigation systems.

CV:

Vasko Sazdovski received the M.Sc. degree in electrical engineering from the Institute of Automation and Systems Engineering, Faculty of Electrical Engineering and Information Technologies, University of Ss. Cyril and Methodius, Skopje, R. Macedonia in 2010 (field of mobile robotics). From 2006 until 2012 studied for a part time PhD at the Centre for Autonomous Systems, Department of Informatics and Systems Engineering, Defence Academy of the United Kingdom, Cranfield University, Swindon, United Kigdom (field of Vision aided Inertial Navigation). His research interests are in robotics, automation and systems engineering. He is particularly interested in simultaneous localization and mapping (SLAM) techniques, sensor and data fusion and estimation methods, mobile robotics, MEMS inertial sensors, Vision/GNSS aided Inertial Navigation and embedded computer systems.