New Breed of Cognitive Robot… a Puppy?
ICT Results (03/21/08)

The European Union-funded COSPAL project has discovered that the
shortcomings of classical rule-based artificial intelligence and artificial
neural networks (ANNs) can be overcome by combining the two approaches, and this
strategy forms the basis of the project's artificial cognitive systems. A team
led by Linkoping University researcher Michael Felsberg has devised a new type
of cognitive, learning robot that employs ANNs to manage the low-level functions
based on the visual input it receives and classical AI to serve as a supervisory
mechanism. "In this way, we found it was possible for the robots to explore the
world around them through direct interaction, create ways to act in it and then
control their actions in accordance," says Felsberg. "This combines the
advantages of classical AI, which is superior when it comes to functions akin to
human rationality, and the advantages of ANN, which is superior at performing
tasks for which humans would use their subconscious, t hings like basic motor
skills and low-level cognitive tasks." The COSPAL robot learned to complete a
shape-sorting puzzle without being provided with specific directions, other than
being told by a human operator when it had performed a correct action or when it
had committed an error. The scalability of the COSPAL researchers' approach is
what distinguishes it the most from what had been the state of the art.
Web Link