Agents
Perceives its environment through sensors and acts upon the environment through actuators
Rational agent: Acts to achieve the best (expected) outcome
For each possible percept sequence, a rational agent should select an action that is expected to maximize its performance measure, given the evidence provided by the percept sequence and whatever built-in knowledge the agent has.
Environment
Classical Planning
Planning: make a sequence of decisions. (choose actions and organize them)
Classical planning deals with Fully observable, Deterministic, Finite, Discrete environments.
Domain independent heuristics
Without good heuristics, planning can become difficult very quickly as the problem size grows.
Problem solving with search takes a lot of time and memory.
Deriving heuristics for individual problems is time consuming.
Domain independent heuristics tend to work well in planning.
Need specific representations:
- Strips
- Satisfiability
- Planning graph