Adaptive Predicates

From AIRWiki
Jump to: navigation, search
Adaptive Predicates
Short Description: Adapting predicates to the robot needs
Coordinator: AndreaBonarini (andrea.bonarini@polimi.it)
Tutor: AndreaBonarini (andrea.bonarini@polimi.it)
Collaborator:
Students: NinoRaffa (TBD@TBD.Com)
Research Area: Robotics
Research Topic: Robot development
Start: 2011/02/25
End: 2012/02/23
Status: Closed
Level: Bs
Type: Thesis

The goal for this project is to analyze how a concept expressed by a fuzzy predicate can be adapted to maintain its semantics while matching the needs to perform an action. An example might be the "Close" predicate to be used by an autonomous wheelchair such as LURCH, whose corresponding fuzzy set might be modified according to the needs, and the actions to be done. For instance, in navigation, up to 80 cm might be "Close", while approaching a table, or entering the door, this limit reduces, although the action triggered as a consequence of the activation of the predicate is the same (e.g. "SPEED is Slow).

This idea is supported by the work on mirror neurons by Rizzolatti, Sinigaglia et al.[1], namely the detection of activation of the same cerebral areas in monkeys when needing to reach an object at different distances in the range or out of the range of arms or of tools in hand.