The Active Learning Environment with Robotics Tangibles (ALERT) system provides
an entertaining and easy to use method of interacting with and programming
a robot. A user simply places a sign in the path of the robot. The robot sees
the sign and obeys the directions given by the sign.
Our robots are Create robots from iRobot
Currently we are using Apple MacBooks and MacBook Pro computers. Smaller, less expensive computers will be used in future generations.
We are using a Unibrain Firewire camera and built in iSight cameras on the MacBooks.
"Reactivision" software is used to enable the computer to recognize the fiducial markers.
The visual markers are included with the Reactivision software package. We have augmented these markers with our own symbols that are human readable. Our marker library includes: left turn, right turn, u-turn, stop, drive slowly, yield (pause), reset, whirlpool (causes the robot to spin for a random time then send it off in a random direction), repeat last command, a lock, a key, etc. We are continuously adding to this library of commands as we develop new scenarios for the robots.
We are using the Roombacomm library from Tod E. Kurt.
This project began, and continues, as a class assignment in the "Active Learning in Mediated Environments" class in the Arts Media and Engineering Program at Arizona State University. The assignment involved the production and testing of a technological learning environment or system.
One of the resources we had available to us to achieve this task was the SMALLab envirnonment. This environment provides camera based motion tracking of up to three objects, large scale video projection on the floor of the space, 4 channel surround sound system and a flexible software infrastructure that facilitates the development of interaction scenarios.
An additional resource at our disposal was several Create robots from iRobot. These robots are easily programmed and can be centrally controlled from a remote host computer using Bluetooth connectivity.
One of the limitations we recognized in the feedback system of SMALLab was its two dimensional and intangible nature. The two current forms of feedback provided in SMALLab are audio and video.
SMALLab audio can be spatialized with sophisitcated software control and this spatialization effectively psychologically extends the space beyond the dimensions of the video projection, but it ultimately remains a fundamentally two dimensional effect. A three dimensional array of speakers have been considered and would expand the dimesionality of the sonic space, but the two dimensional array is the current implemented system.
The second form of feedback in SMALLab is the video projection. The video is projected onto white padded floor tiles to a dimension of approximately 14 x 14'. This allows users to enter the active visual space and be immersed in this interactive environment. The typical SMALLab scenario involves users holding illuminated objects ("glowBalls", "glowSticks" [add link to research poster] and other such interaction objects) which are tracked by a camera based computer vision system. Movements of these objects in the space allows one to interact with the virtual space through a wide variety of gestures. This movement based interaction is highly dimensional and encourages a natural embodied interaction. The visual feedback itself however, remains two dimensional and does make the third dimension more challenging to engage.
Both the audio and visual feedback are non-tactile. One cannot pick up these virtual objects and they cannot affect the physial environment in which they exist (beyond the typical physical effects of low energy light and sound).
Robots provide an opportunity to introduce truly three dimensional, physical feedback into the SMALLab environment. Their actions could be controlled programmatically in the same way that virtual objects could be controlled yet they would have a real physical presence and could physically engage with their environment.
Video projection
Audio feedback
Camera based motion tracking system
Additional sensors and actuators
