Universität Bielefeld - Sonderforschungsbereich 360
PLAYBOT: Towards a Robot for Physically-Disabled Children
Department of Computer Science
University of Toronto, Canada
Wednesday, November 24th, 1999,
10 am, C0-269
The experimental goal of the project is the development of a prototype
environment which will assist disabled children in play, thereby
enhancing their innate capabilities. Imagine a child seated in a robotic
wheelchair. This robot possesses a robotic arm and hand, a stereo-colour
vision robot head, and a communication panel. The communication panel
design is motivated by the BLISS symbolic language, invented by Charles
Bliss in1941 and successfully used by cerebral palsy patients. It is
assumed that the child can point to a picture of a toy (or toys) on the
panel and point to a sequence of pictorially represented actions that
he/she wishes the robot to perform with that toy (or toys). In effect, a
play sentence is specified in terms of a grammatically formed sequence
of actions. The play sequence could involve bringing toys to the child's
table for close inspection and manipulation; it could also involve the
starting or stopping of various automated aspects of the environment
(i.e., motorized trains). The robot would visually locate the toys, plan
the execution of the play sequence, and then together with the child,
move to the proper position and carry out the actions.
Current robots in use for the disabled all rely on the user's intact
visual system to be an integral part of a closed-loop control system.
For example, in one class of robotic aids, specialized sensors are
developed for a controllable muscle the user might have (a finger, eye
brow, eye movement, etc.). The user decides what the robot manipulator
should grasp, and then through a long series of micro-activations of the
robot, visually guides the manipulator to the target object. Each
micro-activation might move a particular joint of a robot arm a small
amount. The key to success here is that the user's visual system senses
the progress of the arm in comparison to the target, and the user's
planning system determines which micro-activation to perform next in
order to move closer to the target. This can be very tedious; the user
may tire easily and the amount of work that might be done in a given
unit of time is very small. Nevertheless, the user does have some
independence and this is important. In the long run, however,
improvements are needed. PLAYBOT's goal is to provide a further step
along the path of enhanced mobility, functionality and independence for
physically disabled children.
Anke Weinberger, 1999-11-16, 1999-11-18