The information needed to describe a robot's trajectory inside a building is not always available, and when it is it is more often than not imprecise, incomplete and unreliable. In order to reason with these data a planner must be endowed with various mechanisms: automatic and incremental learning, reasoning in uncertain contexts, reasoning by default and so on. In our project we study the integration of automatic learning techniques in the trajectory planning module of an Indoor Autonomous Mobile Robot.

Building modeling, in tandem with the interaction control of a robot, make up the core of this application. Our goal is to define the means that will allow an autonomous system to reason inside an environment whose evolution can neither be completely anticipated or controlled.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £60 +P&P)