SegMap: Segment-based mapping and localization using data-driven descriptors

What is it about?

Precisely estimating a robot’s pose in a prior, global map is a fundamental capability for mobile robotics,e.g.autonomous driving or exploration in disaster zones. This task, however, remains challenging in unstructured, dynamic environments, where local features are not discriminative enough and global scene descriptors only provide coarse information. We therefore present SegMap: a map representation solution for localization and mapping based on the extraction of segments in 3D point clouds. Working at the level of segments offers increased invariance to view-point and local structural changes, and facilitates real-time processing of large-scale 3D data. SegMap exploits a single compact data-driven descriptor for performing multiple tasks: global localization, 3D dense map reconstruction, and semantic information extraction.

Featured Image

Photo by Lenin Estrada on Unsplash

Photo by Lenin Estrada on Unsplash

Why is it important?

Mapping and localization are fundamental competencies for mobile robotics and have been well-studied topics over the last couple of decades (Cadena et al. (2016)). Being able to map an environment and later localize within it unlocks a multitude of applications, that include autonomous driving,rescue robotics, service robotics, warehouse automation or automated goods delivery, to name a few. Robotic technologies undoubtedly have the potential to disrupt those applications within the next years. In order to allow for the successful deployment of autonomous robotic systems in such real-world environments, several challenges need to be overcome: mapping, localization and navigation in difficult conditions, for example crowded urban spaces, tight indoor areas or harsh natural environments. Reliable, prior-free global localization lies at the core of this challenge. Knowing the precise pose of a robot is necessary to guarantee reliable,robust and most importantly safe operation of mobile platforms and also allows for multi-agent collaborations.