Model-based Reflex Agent

An Agent (similar to Simple Reflex Agent) whose Agent Function depends on a world model that consists of:

• A Belief State (this is the different part where we keep track of the state of the world instead of only looking at the last Percept)
• A Sensor Model
• A Transition Model
  • maps states to actions
• an Agent Function
  • maps states to actions iteratively based on the sensor models output

General Agent Program

function Model−Based−Agent (percept) returns an action
		var state /∗ a description of the current state of the world ∗/ persistent rules /∗ a set of condition−action rules ∗/
	var action /∗ the most recent action, initially none ∗/

	state := Update−State(state,action,percept)
	rule := Rule−Match(state,rules)
	action := Rule−action(rule)
	return action

Schema

Problems

• does not always determine what to do (in a Rational way)

Further

→ Goal-based Agent