Learning to Encode Vision on the Fly in Unknown Environments: A Continual Learning SLAM Approach for Drones

• Published in: IEEE International Symposium on Safety, Security and Rescue Robotics pp. 373 - 378
• Main Authors: Safa, Ali, Verbelen, Tim, Ocket, Ilja, Bourdoux, Andre, Sahli, Hichem, Catthoor, Francky, Gielen, Georges G.E.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.11.2022
• Subjects: Deep learning; Safety; Security; Simultaneous localization and mapping; Surveillance; Training; Visualization
• ISSN: 2475-8426
• DOI: 10.1109/SSRR56537.2022.10018713

Learning to safely navigate in unknown environ-ments is an important task for autonomous drones used in surveillance and rescue operations. In recent years, a number of learning-based Simultaneous Localisation and Mapping (SLAM) systems relying on deep neural networks (DNNs) have been proposed for applications where conventional feature descriptors do not perform well. However, such learning-based SLAM systems rely on DNN feature encoders trained offline in typical deep learning settings. This makes them less suited for drones deployed in environments unseen during training, where continual adaptation is paramount. In this paper, we present a new method for learning to SLAM on the fly in unknown environments, by modulating a low-complexity Dictionary Learning and Sparse Coding (DLSC) pipeline with a newly proposed Quadratic Bayesian Surprise (QBS) factor. We experimentally validate our approach with data collected by a drone in a challenging warehouse scenario, where the high number of ambiguous scenes makes visual disambiguation hard.