Abstract
Indoor navigation has gained significant importance in the last decade due to its applications in the location based services industry. The outdoor navigation technology of GNSS by itself does not offer good solutions due to presence of multipath and low SNR in indoor environments. Complimentary sensors like IMU, magnetometers, cameras, etc., have their own set of problems. A camera however, provides a wealth of information for position estimation.
The main aim of this thesis is to analyse camera based navigation systems using stationary figures of opportunity. Unlike traditional navigation systems that utilise coded markers, this thesis performs navigation based on planar stationary figures that are inherently present in the environment, hence the more structure the more information. The novelty of this thesis lies in its usage of Hough transforms to do the same. Initial experimental results demonstrate that an average webcam quality camera provides enough information to extract reasonably good navigation coordinates under specific conditions.
