Swarm Robotics - North Carolina State University

chemistoddΤεχνίτη Νοημοσύνη και Ρομποτική

6 Νοε 2013 (πριν από 3 χρόνια και 10 μήνες)

120 εμφανίσεις

The WolfBot:

Swarm Robotics Platform with Image
Processing Applications


Jimit Patel

Dr. Edgar
Lobaton


Department of Electrical and Computer Engineering

North Carolina State University

What is Swarm Robotics?




Multi
-
robot system inspired from social
insects



Ability to work co
-
operatively to achieve a
common goal



“Swarm Intelligence”: simple set of rules for
individuals, sophisticated collective behavior
for the group.



Research in Swarm Robotics


Communications


Control Approach


Mapping and localization


Learning and task allocation


Reconfigurable Robotics


Object transportation and manipulation


Applications


Reconnaissance scenarios such as a
natural disaster, search and rescue
missions, surveillance, security purposes,
mapping unknown terrain or distributed
sensing …



Gathering information about environment


Computer Vision

Existing Swarm Robots


Trade
-
off between cost, size and features



Very few swarm robots have camera


From those which have cameras, ability to
take images at
good

resolution is rare


Those which do take images at good resolution,
often lack ability to process images


Finally, those which have ability to process images, aren’t fast
enough! (or are very expensive)


Comparison with different swarm
robots

Name(s)
Cost per unit
Camera
Kilobot, R-one, Elisa III, Alice, Libot,
Robomote, I-swarm, Jasmine, i-robot, …
< $350
NA
E-puck
$730
40x40 @ 4 FPS
MM-robot[1]
NA
VGA @ 20 FPS max
Khepera III
$3,500
HD, logitech
Corobot
$4,000
VGA
The WolfBot


Low Cost


High Performance


Designed for swarm robotics


Ability for on
-
board image processing


Easily replaceable camera and
communication modules


Omnidirectional drive






Processor


Beaglebone

-

TI AM3359


ARM Cortex A8, 720 MHz, 256 MB DDR2
RAM





-

Ubuntu 12.10
armhf





-

OpenCV

2.4.2









C



Sensors


Accelerometer, Magnetometer


Ambient Light sensors


IR Distance Measurement sensors



Camera


MS LifeCam HD3000



Communication


IR (Robot
-
to
-
Robot)


Zigbee


Wifi

(video stream/ftp)









Sensors


LSM303DLHC 3D Accelerometer and digital
compass



Ambient Light Sensors



IR Distance Measurement sensors (~10cm
to 100cm)



Microsoft LifeCam HD
-
3000



720p
HD, 50 FPS


Communications


Nano
wi
-
fi

adaptor used for WLAN. Used
for video stream/ftp




IR Transceiver Beacon for short range,
line of sight communication (range
upto

15
ft
)



Zigbee

communication for mesh network
in the swarm



Battery and Power Management




7.4V Lithium Ion 5200mAh
Battery



Two TPS5420 DC
-
DC Converters to step
down the voltage to 5V and 3.3V




Experimental Results


Currently, the
Wolfbot

is programmed to
receive its co
-
ordinates from
OptiTrack

(local positioning system) and move in
random directions till it detects an object.



Peak current draw of
~1.5A
and
upto

2.5
hrs

of continuous operation (motion,
sensing and
video stream at 720p
resolution)





Edge Detection


Used
cvCanny
() from
OpenCV

on 512x512
pixels image


Execution time : 0.05 sec (at 500MHz)*









* For comparison, the execution time is 0.34sec for CITRIC platform which used IPP
canny edge detection at 520MHz for 512x512 pixel images. Same program on
intel

core
-
i5 laptop takes 0.008 sec at 2.6 GHz



Face Detection


Used

Haar
-
Cascades for face
detection from
OpenCV

on 512x512 pixels image










Summary



Introduction of image processing
capabilities in swarm robotics research



Low cost design (
approx

$500)



Facilitates experiments on different
research domains in swarm intelligence




References

1)
Haverinen
, J.,
Parpala
, M., &
Roning
, J. (2005). A Miniature Mobile Robot With a Color Stereo Camera System
for Swarm Robotics Research.
IEEE International Conference on Robotics and Automation (ICRA 2005)
, (April 18
-
22), 2494

2497
.

2)
Rubenstein
, M.,
Ahler
, C., &
Nagpal
, R. (2012).
Kilobot
: A low cost scalable robot system for collective
behaviors.
2012 IEEE International Conference on Robotics and Automation
, 3293

3298.

3)
Zahugi
, E. M. H.,
Shabani
, A. M., & Prasad, T. V. (2012).
Libot

: Design of a Low Cost Mobile Robot for Outdoor
Swarm Robotics.
IEEE International Conference on Cyber Technology in Automation, Control and Intelligent Systems
,
(May 27
-
31), 342

347.

4)
Sibley, G. T.,
Rahimi
, M. H., &
Sukhatme
, G. S. (2002).
Robomote

: A Tiny Mobile Robot Platform for Large
-
scale
Ad
-
hoc Sensor Networks


’.
IEEE International Conference on Robotics and Automation
, (May), 1143

1148.

5)
Mondada
, F.,
Bonani
, M.,
Raemy
, X., Pugh, J.,
Cianci
, C.,
Klaptocz
, A.,
Zufferey
, J., et al. (2006). The e
-
puck , a
Robot Designed for Education in Engineering
.

6)
Blazovics
, L.,
Varga
, C.,
Csorba
, K.,
Feher
, M.,
Forstner
, B., &
Charaf
, H. (2011). Vision Based Area Discovery
with Swarm Robots.
2011 Second Eastern European Regional Conference on the Engineering of Computer Based
Systems
, 149

150. doi:10.1109/ECBS
-
EERC.2011.32