Single Vision-Based Self-Localization for Autonomous Robotic Agents

Marios Avgeris, Dimitrios Spatharakis, Nikolaos Athanasopoulos, Dimitrios Dechouniotis, Symeon Papavassiliou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)
102 Downloads (Pure)

Abstract

We present a single vision-based, self-localization method for autonomous mobile robots in a known, indoor environment. This absolute localization method is landmark assisted, therefore, we propose an algorithm that requires the extraction of a single landmark feature i.e., the length of a known edge. Our technique is based on measuring the distance from two distinct, arbitrarily
positioned landmarks in the robot’s environment, the locations of which are known a priori. A single camera vision system is used to perform distance estimation. The developed framework is applied to tracking a robot’s pose,
i.e., its position and orientation, in a Cartesian coordinate system. The position of the robot is estimated using a bilateration method, while its orientation calculation utilizes tools from projective geometry. The validity and feasibility
of the approach are demonstrated through experiments.
Original languageEnglish
Title of host publication2019 7th International Conference on Future Internet of Things and Cloud Workshops (FiCloudW)
Publisher IEEE
Number of pages7
ISBN (Electronic)978-1-7281-4411-5
DOIs
Publication statusPublished - 02 Apr 2020

Fingerprint

Dive into the research topics of 'Single Vision-Based Self-Localization for Autonomous Robotic Agents'. Together they form a unique fingerprint.

Cite this