English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

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

Learning to SLAM on the Fly in Unknown Environments: A Continual Learning Approach for Drones in Visually Ambiguous Scenes