Self-Supervised Visual Place Recognition Learning in Mobile Robots
This addresses the problem of drift correction in mobile robot navigation, but it is incremental as it builds on existing methods like CNN models and GPS-aided solutions.
The paper tackles visual place recognition for robot navigation by developing a self-supervised approach that casts loop-closure identification as a metric learning problem, using GPS-aided data to bootstrap labels and learn a distance metric for image descriptors, resulting in improved disambiguation of visual scenes.
Place recognition is a critical component in robot navigation that enables it to re-establish previously visited locations, and simultaneously use this information to correct the drift incurred in its dead-reckoned estimate. In this work, we develop a self-supervised approach to place recognition in robots. The task of visual loop-closure identification is cast as a metric learning problem, where the labels for positive and negative examples of loop-closures can be bootstrapped using a GPS-aided navigation solution that the robot already uses. By leveraging the synchronization between sensors, we show that we are able to learn an appropriate distance metric for arbitrary real-valued image descriptors (including state-of-the-art CNN models), that is specifically geared for visual place recognition in mobile robots. Furthermore, we show that the newly learned embedding can be particularly powerful in disambiguating visual scenes for the task of vision-based loop-closure identification in mobile robots.