Abstract

Robots must perform tasks efficiently and reli- ably while acting under uncertainty. One way to achieve efficiency is to give the robot common- sense knowledge about the structure of the world. Reliable robot behaviour can be achieved by mod- elling the uncertainty in the world probabilistically. We present a robot system that combines these two approaches and demonstrate the improvements in efficiency and reliability that result. Our first con- tribution is a probabilistic relational model integrat- ing common-sense knowledge about the world in general, with observations of a particular environ- ment. Our second contribution is a continual plan- ning system which is able to plan in the large prob- lems posed by that model, by automatically switch- ing between decision-theoretic and classical proce- dures. We evaluate our system on object search tasks in two different real-world indoor environ- ments. By reasoning about the trade-offs between possible courses of action with different informa- tional effects, and exploiting the cues and general structures of those environments, our robot is able to consistently demonstrate efficient and reliable goal-directed behaviour.