UNCERTAIN ROBOT ENVIRONMENT MODELING USING FUZZY NUMBERS

In this paper, we present a fuzzy-oriented methodology to model an unc ertain geometric robot environment and to manipulate geometric uncerta inties between robot coordinate frames. We describe any geometric prim itive of robot environment as a parameter vector in parameter space. N ot only ill-known values of the parameterized geometric primitives but the uncertain quantities of coordinate transformations are represente d by means of fuzzy numbers restricted to appropriate membership funct ions. For consistent interpretation about geometric primitives between different coordinate frames, we manipulate these uncertain quantities using fuzzy arithmetic. As an application, we deal with a 2-dimension al straight line correspondence problem of a mobile robot vision syste m. This example shows the usefulness of the proposed methodology.