Information Technology Reference
In-Depth Information
5 Summary and Future Work
A key issue in robotics is that the amount of software necessary for even the
most basic competency is quite large. Framework TiViPE has been extended
with robotics library to substantially reduce the amount of coding necessary to
control a robot.
The robot control platform is aimed to develop to an end-user platform.
So far, a user-friendly architecture has been developed, which makes possible
non-specialists to co-develop useful behaviors together with a person with some
knowledge in programming. The future user of the robots is often not a program-
mer or a person with keen interest in robotics, hence we need to provide easy
to use building blocks for these users to provide full control over a robot. This
is done by providing ready graphical components for sensing the internal values
of a robot as well as video and auditory modules. Further a single component
is used to command the robot. No more than a single module for commanding
a robot is required since a textual robotics language has been introduced with
an a-priori known list of commands. This yields a simple but powerful language
to gain full control over a robot. In this framework constructing scenario's has
become an interplay between reading out sensory information and based upon
this information generating a command string to control a robot. Creating these
blocks have become entirely trivial, andinthatrespectwehaveshiftedthe
complexity to creating scenario's where numerous blocks are involved.
So far the framework has been tested by design students and professionals
that have created behaviors within TiVIPE [14,1] and image analysis for robot
control [15,16]. Our future focus will be on scenario's and a layered component
infrastructure, here the essence is to compound a set of blocks into meaningful
partial sub-scenario's. Compounding itself is done without any addition program-
ming with graphical programming environment TiViPE. Hence, once these sub-
scenarios are created it will be easy for users to combine several sub-scenarios.
For the particular application we are working on, we envision that therapists can
do this to create training scenarios for autistic children. The scopes of application
of such a generic framework, however, extends from exact control to applications
in behavioral robotics and social robotics where issues of uncertainties need to
be addressed.
Acknowledgments
We gratefully acknowledge the support of WikiTherapist project by the
Innovation-Oriented Research Programme 'Integrated Product Creation and Re-
alization (IOP IPCR)' of the Netherlands Ministry of Economic Affairs.
We also thank Ph.D. students, J. Gillesen and S. Boere, both supported by
the WikiTherapist project, for developing novel robot behavior modules within
TiViPE.
Search WWH ::
Custom Search