Lazy Sequences Matching Under Substantial Appearance Changes (Short Paper)

The ability to localize in changing environments is essential for robust long-term navigation. Robots operating over extended periods of time must be able to handle substantial appearance changes. In this paper, we investigate the problem of efficiently coping with seasonal changes in online localization. We propose an online lazy data association approach for matching streams of incoming images to a reference image sequence. We propose a search heuristic to quickly find matches between the current image sequence and the database. We present an experimental evaluation using real world data containing substantial seasonal changes and show that our approach can efficiently match sequences by requiring comparably small number of image comparisons.