Title: Online map fusion system based on sparse point-cloud

Authors: Shiqin Sun; Benlian Xu

Addresses: School of Electrical and Automatic Engineering, Changshu Institute of Technology, Changshu, China ' School of Mechanical Engineering, Changshu Institute of Technology, Changshu, China

Abstract: With the gradual maturity of single-robot simultaneous localisation and mapping (SLAM) technology, the idea of using a robotic team to perform this task has attracted more and more attention. In this paper, we proposed an online map fusion strategy for a centralised architecture, in which a two-robot map building system employs multiple independent robots to work as agents and each being able to explore the environment independently through its own SLAM algorithm with a camera sensor and a central control module. When implementing fusion, the local map information is packaged and sent to the server. The server is responsible for map fusion, optimisation and returning the global map to each agent. This makes each agent incorporate observations of others timely in its own SLAM running. Under the proposed framework, a vision-based SLAM algorithm is employed and tested, and the results verify that the strategy is suitable for multi-robot scenarios.

Keywords: simultaneous localisation and mapping; SLAM; map building; robot navigation; online map fusion.

DOI: 10.1504/IJAAC.2021.116424

International Journal of Automation and Control, 2021 Vol.15 No.4/5, pp.585 - 610

Received: 28 Nov 2019
Accepted: 04 Jan 2020

Published online: 23 Jul 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article