In this paper, we propose a definition of goal achievability: given a
basic action theory describing an initial state of the world and some
primitive actions available to a robot, including some actions which r
eturn binary sensing information, what goals can be achieved by the ro
bot? The main technical result of the paper is a proof that a simple r
obot programming language is universal, in that any effectively achiev
able goal can be achieved by getting the robot to execute one of the r
obot programs. The significance of this result is at least twofold. Fi
rst, it is in many ways similar to the equivalence theorem between Tur
ing machines and recursive functions, but applied to robots whose acti
ons are specified by an action theory. Secondly, it provides formal ju
stifications for using the simple robot programming language as a foun
dation for our work on robotics. (C) 1998 Elsevier Science B.V. All ri
ghts reserved.