The navigation system of an unmanned mobile platform in different environmental conditions

Currently, the importance of using unmanned mobile platforms in various industries and production is increasing. This has led to an increase in the need for accurate positioning and safe use of these platforms. The main goal of this research is to develop a vision-based navigation system that is invariant to changes in lighting conditions. For robots that navigate using only cameras, changes in illumination can cause failures during re-localization during autonomous navigation. This article describes a multi-session visual simultaneous localization and mapping (SLAM) approach that allows for the creation of a map consisting of multiple variations of the same location in different lighting conditions. This map can be used any time of day to improve localization both indoors and outdoors, making it suitable for various applications.

Authors: E. A. Maslennikova, A. Salamakhin, D. M. Filatov, A. A. Truts, P. S. Grigoriev

Direction: Informatics, Computer Technologies And Control

Keywords: robotics, mobile platform, stereo camera, navigation, localization, visual markers, control system, autonomous control

---

View full article