In this work, we develop a feature-based real-time Visual-Inertial-Odometric system that operates in an indoor environment for Ground Mobile Robots. The system is based on the opensource ORB-SLAM system, and it is optimized to work better for ground mobile robots by adding SE2-XYZ constraints, which constraint the movement along the Z-axis.
Ха Т., Махмуд Ж. (науч. рук. Колюбин С.А.) Visual-Inertial-Odometric Localization and Mapping for Indoor Ground Mobile Robot using SE2-XYZ Constraints // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2021]. URL: https://kmu.itmo.ru/digests/article/6289