Vision Based Environment Mapping By Network Connected Multi-Robotic System.

AHMED, M Shuja, SAATCHI, Reza and CAPARRELLI, Fabio (2013). Vision Based Environment Mapping By Network Connected Multi-Robotic System. In: Proceedings of the 3rd International Conference on Pervasive Embedded Computing and Communication Systems - Volume 1: PECCS. SCITEPRESS, 49-54.

[img]
Preview
PDF
PaperID41.pdf - Accepted Version
All rights reserved.

Download (757kB) | Preview
Official URL: http://www.scitepress.org/digitallibrary/Publicati...
Link to published version:: https://doi.org/10.5220/0004314600490054

Abstract

The conventional environment mapping solutions are computationally very expensive and cannot effectively be used in multi-robotic environment, where small size robots with limited memory and processing resources are used. This study provides an environment mapping solution in which a group of small size robots extract simple distance vector features from the on-board camera images. The robots share these features between them using a wireless communication network setup in infrastructure mode. For mapping the distance vector features on a global map and to show a collective map building operation, the robots needed their accurate location and heading information. The robots location and heading information is computed using two ceiling mounted cameras, which collective localises the robots. Experimental results show that the proposed method provides the required environmental map which can facilitate the robot navigation operation in the environ- ment. It was observed that, using the proposed approach, the near by object boundaries can be mapped with higher accuracy comparatively the far lying objects.

Item Type: Book Section
Research Institute, Centre or Group: Materials and Engineering Research Institute > Engineering Research
Identification Number: https://doi.org/10.5220/0004314600490054
Depositing User: Fabio Caparrelli
Date Deposited: 20 Apr 2018 14:48
Last Modified: 20 Apr 2018 14:58
URI: http://shura.shu.ac.uk/id/eprint/13710

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics