Camera trajectory estimation for indoor robot odometry using stereo images and inertial measurements


Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Makina Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2016

Öğrenci: ANIL HORASAN

Danışman: MELİK DÖLEN

Özet:

In this study, the development and implementation of an algorithm for stereo visual-inertial odometry are described. The study spans the complete process from analyzing the sensory data to the development of a robot odometry algorithm. The criteria for indoor visual-inertial odometry include using low-cost sensor systems, having an error less than five percent of the movement regardless of the distance covered, and building a robust algorithm in the presence of geometric and photometric invariances as well as noise. Utilizing the complementary characteristics of two different sensors, these steps are followed: First, orientation, velocity and position are estimated using inertial measurements. Second, the machine vision algorithm is developed consisting of feature detection and extraction, feature tracking in consecutive images, disparity map calculation, outlier rejection, motion estimation and optimization. Finally, inertial estimates are fused to visual pose estimates using EKF and a proposed filter. In this research, all the algorithms are implemented offline and tested using EuRoC MAV datasets. The results show that it is possible to achieve less than five percent positional errors in different indoor environments.