A decision theoretic problem.

The policy is a sequential composition of

• agent actions,


• senseEffect(a) sensing actions (they gather information about a real outcome of a stochastic action a) and


• conditionals ( ), where is a situation calculus formula that provides a test condition to decide which branch of a policy the agent can take given the result of sensing.

Example:

pickup(Coffee,55) ; startGo(MO,office(Ann),55) ; endGo(MO,office(Ann),90) ;