Authors:
Anssi Kemppainen
;
Janne Haverinen
and
Juha Röning
Affiliation:
University of Oulu, Finland
Keyword(s):
Distributed sensing, relative pose estimation, multi-robot system, formation control.
Related
Ontology
Subjects/Areas/Topics:
Informatics in Control, Automation and Robotics
;
Mobile Robots and Autonomous Systems
;
Robotics and Automation
Abstract:
Distributed sensing refers to measuring systems where, instead of one sensor, multiple sensors are spatially distributed to improve the robustness of the system, increase the relevancy of the measurements and cut costs, since smaller and less precise sensors are used. Spatially distributed sensors fuse their measurements into the same co-ordinates, which requires the relative positions of the sensors. In this paper we present a distributed multi-robot sensing system in which the relative poses (positions and orientations) of the robots are estimated using an infrared location system. The relative positions are estimated using intensity and bearing measurements of received infrared signals. The relative orientations are obtained by fusing the position estimates of the robots. The location system enables a group of robots to perform distributed and co-operative environment sensing by maintaining a given formation while the group measures distributions of light and a magnetic field, for
example. In the experiments, a group of three robots moved and collected spatial information (i.e. illuminance and compass headings) from the given environment. The information was stored on grid maps that present illuminance and compass headings. The experiments demonstrated the feasibility of using the distributed multi-robot sensing system in mobile sensing applications.
(More)