ON INFORMATION INVARIANTS IN ROBOTICS

We consider the problem of determining the information requirements to perform robot tasks, using the concept of information invariants. Thi s paper represents our attempt to characterize a family of complicated and subtle issues concerned with measuring robot task complexity. We also provide a first approximation to a purely operational theory that addresses a narrow but interesting special case. We discuss several m easures for the information complexity of a task: (a) How much interna l state should the robot retain? (b) How many cooperating agents are r equired, and how much communication between them is necessary? (c) How can the robot change (side-effect) the environment in order to record state or sensory information to perform a task? (d) How much informat ion is provided by sensors? and (e) How much computation is required b y the robot? We consider how one might develop a kind of ''calculus'' on (a)-(e) in order to compare the power of sensor systems analyticall y. To this end, we attempt to develop a notion of information invarian ts. We develop a theory whereby one sensor can be ''reduced'' to anoth er (much in the spirit of computation-theoretic reductions), by adding , deleting, and reallocating (a)-(e) among collaborating autonomous ag ents.