Vision-based Docking of a Mobile Robot

EasyChair Preprint 5092

Abstract

For mobile robots to be considered autonomous they must reach target locations in required pose, a procedure referred to as docking. Popular current solutions use LiDARs combined with sizeable docking stations but these systems struggle by incorrectly detecting dynamic obstacles. This paper instead proposes a vision-based framework for docking a mobile robot. Faster R-CNN is used for detecting arbitrary visual markers. The pose of the robot is estimated using the solvePnP algorithm relating 2D-3D point pairs. Following exhaustive experiments, it is shown that solvePnP gives systematically inaccurate pose estimates in the x-axis pointing to the side. Pose estimates are off by ten to fifty centimeters and could therefore not be used for docking the robot. Insights are provided to circumvent similar problems in future applications.

Keyphrases: PnP problem, bounding box, camera pose estimation, cnn based object detector, computer vision, deep learning, mobile robot docking, object detection, path planning, pose estimation, preprint arxiv, solvepnp algorithm

@booklet{EasyChair:5092,
  author    = {Andreas Kriegler and Wilfried Wöber},
  title     = {Vision-based Docking of a Mobile Robot},
  howpublished = {EasyChair Preprint 5092},
  year      = {EasyChair, 2021}}