Indoor MAV Auto-Retrieval Using Fast 6D Relocalisation

Jose Martinez-Carranza, Walterio Mayol-Cuevas

Research output: Contribution to journalArticle (Academic Journal)peer-review

8 Citations (Scopus)
499 Downloads (Pure)


This paper develops and evaluates methods for performing auto-retrieval
of a MAV using fast 6D relocalisation from visual features. Auto-retrieval
involves a combination of guided operation to direct the vehicle through
obstacles using a human pilot and autonomous operation to navigate the
vehicle on its return or during re-exploration. This approach is useful in
tasks such as industrial inspection and monitoring, and in particular to
operate indoors in GPS-denied environments. Our relocalisation methodology
contrasts two sources of information: depth data and feature covisibility
but in a novel manner that validates matches before a RANSAC
procedure. The result is the ability of performing 6D relocalisation at an
average of 50Hz on individual maps containing 120K features. The use of
feature co-visibility reduces memory footprint as well as removes the need
to employ depth data as used in previous work. This paper concludes
with an example of an industrial application involving visual monitoring
from a MAV aided by autonomous navigation.
Original languageEnglish
Pages (from-to)119-130
Number of pages12
JournalAdvanced Robotics
Issue number2
Publication statusPublished - 23 Oct 2015


  • MAV
  • aerial robotics
  • autonomous navigation
  • visual odometry
  • relocalisation


Dive into the research topics of 'Indoor MAV Auto-Retrieval Using Fast 6D Relocalisation'. Together they form a unique fingerprint.

Cite this