![](bg4.png)
in the mosaics.
In experiments, it was noticed that, if not enough
matching features are found between the consecutive
images, then erroneous stitching of the images can
occur. An example is shown in Figure 7a. In the
beginning, the images were stitched properly. The
problem occurred when the last two images shown in
Figures 7b and 7c were stitched. The information
contributed by these images is identified by the blue
arrow. When these two images are compared with
the mosaic, it can be noticed that they are stitched at
wrong points. The two correct corresponding points
where the image stitching should be performed, are
identified with red arrow. Although there is sufficient
overlap between these images, there are no objects in
this overlapping region. This causes reduced match-
ing features between these images and false homog-
raphy was computed.
Figure 7: (a) Erroneous Stitching in Mosaic. (b) Second
Last Image for Mosaic. (c) Last Image for Mosaic.
4 CONCLUSIONS
In this study, a distributed mosaic formation and ob-
ject detection approach in a multi-processor robot was
presented. The overall task was distributed among
three processing modules. This distributed implemen-
tation enables the master processing module to focus
on the robot locomotion task as it can process the
images at faster rate. At the same time, the master
module utilises the processing resources of the slave
robots to perform the computationally expensive task,
that is mosaic generation and object detection. During
the experiments, it was observed that, if small number
of objects are present on the location where a robot
tries to generate mosaics, then erroneous stitching of
the images is expected. The reason for this was the
lack of common features between the two consecu-
tive images. To overcome this problem, the use of a
compass in the robot can also be made.
ACKNOWLEDGEMENTS
Funded by EU-FP7 research project REPLICATOR.
REFERENCES
Brugali, D. and Fayad, M. (2002). Distributed computing in
robotics and automation. In Transactions on Robotics
and Automation, Vol. 18, No.4, Pages: 409-420.
Defago, X. (2001). Distributed computing on the move:
From mobile computing to cooperative robotics and
nano robotics. In In Proc. 1st ACM Intl Workshop on
Principles of Mobile Computing, Pages 49-55.
Fukuda, T. and Nakagawa, S. (1988). Dynamically recon-
figurable robotic system. In Proc. of Intl. Conf. On
Robotics and Automation, Vol.3, Pages:1581-1586.
Kernbach, S., Hamann, H., and Stradner, J. (2009). On
adaptive self-organization in artificial robot organ-
isms. In The First International Conference on
Adaptive and Self-adaptive Systems and Applications
(ADAPTIVE), Pages: 33-43.
Kernbach, S., Szymanski, M., Schmickl, T., and Corradi, P.
(2008). Symbiotic robot organisms: Reprlicator and
symbrion projects. In PerMIS, Special Session on EU-
projects, Pages: 62-69.
Kumar, V., Bekey, G., and Sanderson, A. (2006). Chap-
ter 7 networked robots, assessment of international re-
search and development in robotics, pages: 73-80. In
http://www.wtec.org/robotics/. NASA.
Replicator (2008). Robotic evolutionary self-programming
and self-assembling organisms. In 7th Framework
Programme Project No FP7-ICT-2007.2.1. European
Communities, URL: http://symbrion.org/.
Tuci, E., Gro, R., Trianni, V., Mondada, F., Bonani, M.,
and Dorigo, M. (2006). Cooperation through self-
assembling in multi-robot systems. In ACM Trans-
actions on Autonomous and Adaptive Systems, Vol.1,
No.2, Pages: 115-150.
Yim, M., Shen, W., Salemi, B., Rus, D., Moll, M., Lipson,
H., Klavins, E., and Chirikjian, G. (2007). Modular
self-reconfigurable robot system. In Robotics and Au-
tomation Magazine, Vol.14, No.1, Pages:43-52.
Zhang, Y., Yim, M., Eldershaw, C., Duff, D., and Roufas,
K. (2003). Scalable and reconfigurable configurations
and locomotion gaits for chain-type modular reconfig-
urable robots. In International Symposium on Com-
putational Intelligence in Robotics and Automation,
IEEE, Vol.2, Pages: 893-899.
PECCS2013-InternationalConferenceonPervasiveandEmbeddedComputingandCommunicationSystems
138