IEEE Robotics & Automation Magazine - December 2013 - 68

eye-bots provide a fast and systematic exploration
of the environment, whereas foot-bots provide
longer-term storage of exploration information on
the ground. Whenever an exploring eye-bot finds
Eye-Bot
Foot-Bot
Swarm Search
Chain
a shelf containing objects, it communicates
knowledge of its discovery back to the deployShelf Found
ment area through the heterogeneous network of
Position
eye-bots and foot-bots.
Foot-Hand-Bot
Hand-Bot
of Object
Location
The swarmanoid now needs hand-bots at
Self-Assemble Object
Retrieve Object
on Shelf
of Shelf/
the shelf location to retrieve the objects. In the
Deployment
Retrieved
deployment area, foot-bots thus assemble to
Vertically Aligned
Area
Assembled
hand-bots (Foot-Hand-Bot Self-Assemble)
with Object
Shelf Reached
and start collectively transporting them to the
Foot-Hand-Bot
Foot-Hand-Bot
shelf [36]. We refer to the composite entity
Navigate
Align and Disassemble
formed by the foot-bots assembled to a handbot as a foot-hand-bot [see Figure 1(a)].
Deployment Area
Reached
Guided by the foot-bot chain, the foot-handbots can navigate through the environment,
Foot-Hand-Bot
following the shortest path from the deployDisassemble and Reset
ment area to the shelf (Foot-Hand-Bot
Navigate). When the foot-hand-bot arrives at
Figure 8. Search-and-retrieval behavioral components. The dotted lines
a shelf location, the eye-bot that found the
represent information transfer. The solid lines represent handing over of control
shelf conveys information about the 3-D locato another behavioral component. The swarmanoid must initially be launched
tion of a particular object on the shelf to the
by executing the heterogeneous exploration and communication network
behavioral components, while each execution of the foot-hand-bot disassemble
foot-hand-bot. The information tells the footand reset behavior marks the successful return of a single object to the
hand-bot's constituent hand-bot where it
deployment area.
should climb and to what height. Based on
this information, the foot-hand-bot aligns
itself vertically with the object it will retrieve. Then the
Shelf with Object
foot-bots disassemble from the hand-bot and act as
Deployment
to Be Retrieved
Area
markers to subsequent foot-hand-bots letting them know
not to approach the shelf at that location (Foot-Hand-Bot
Corridor
Large
Align and Disassemble). The hand-bot climbs the strucRoom
ture, grasps the object, and descends from the shelf
(Hand-Bot Retrieve Object). The foot-hand-bot then
reassembles (Foot-Hand-Bot Self-Assemble) and follows
the foot-bot chain back to the deployment area (FootHand-Bot Navigate). Once the object has been successfully returned to the deployment area, the foot-hand-bot
disassembles, and its constituent robots wait until they
are needed for further swarmanoid tasks (Foot-Hand-Bot
(a)
Disassemble and Reset).
Heterogeneous Exploration and Communication Network

Shelf with Object
to Be Retrieved
Large Room
3.4 m 12.5 m
Deployment Area
Corridor
1.85 m

1m
17 m
(b)

Figure 9. A drawing of the test arena for the swarmanoid
demonstration. (a) The parallel projection. (b) The floorplan.

IEEE ROBOTICS & AUTOMATION MAGAZINE

DECEMBER 2013

Search and Retrieval: A Real-World Demonstration
We demonstrated our integrated search-and-retrieval
behavior in a real-world scenario. Our experiment
involved a real-world instantiation of the generic searchand-retrieval task in an environment containing a single
shelf and book. The arena we used can be seen in Figure 9.
We successfully demonstrated that a swarmanoid with no
a priori knowledge of the environment was able to find the
shelf and retrieve the book. This scenario integrated various swarmanoid abilities, ranging from task allocation to
collective search, from self-assembly to cooperative transport, from object retrieval to cooperative navigation in
complex environments.

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2013