Implementing belief change in the situation calculus and an application

Abstract

Accounts of belief and knowledge in the Situation Calculus have been developed and discussed for some time yet there is no extant implementation. We develop a practical implementation of belief and belief change in the Situation Calculus based on default logic for which we have an implemented solver. After establishing the mapping with default logic we demonstrate how belief change in the Situation Calculus can be used to solve an interesting problem in robotics - reasoning with misleading information. Motivated by a challenge in the RoboCup@Home competition, we give a solution to the problem of planning robustly in cases where operators provide the robot with misleading or incorrect information. © 2013 Springer-Verlag.