Navigation of robotics platform using monocular visual odometry

Abstract

Precise navigation of mobile robots in unknown areas is common problematic in many research papers. For navigation in outdoor areas Global Navigation Satellite System (GNSS) is typically used. For better position estimation in Europe augmentation satellite system called European Geostationary Navigation Overlay Service (EGNOS) it is used. In US it is similar system called Wide Area Augmentation System (WAAS). These systems are not usable in indoor areas because the weak of signal and the low position accuracy. For indoor areas, it is necessary to use an alternative navigation system - independent on satellite navigation. In our research we are focusing on visual odometry using a monocular camera. Advanced image processing methods on camera images to achieved accurate visual odometry are applied.