Intelligent Agents

An Intelligent Agent perceives it environment via sensors and acts rationally upon that environment with its effectors (actuators). Hence, an agent gets percepts one at a time, and maps this percept sequence to actions.

Properties of the agent

• Autonomous
• Interacts with other agents plus the environment
• Reactive to the environment
• Pro-active (goal- directed)

What do you mean, sensors/percepts and effectors/actions?

For Humans

• Sensors: Eyes (vision), ears (hearing), skin (touch), tongue(gestation), nose (olfaction), neuromuscular system (proprioception)
• Percepts:
  • At the lowest level – electrical signals from these sensors
  • After preprocessing – objects in the visual field (location, textures, colors, …), auditory streams (pitch, loudness, direction)
• Effectors: limbs,digits, eyes, tongue
• Actions: lift a finger, turn left, walk, run, carry an object

The Point: percepts and actions need to be carefully defined, possibly at different levels of abstraction

A more specific example: Automated Taxi Driving System

• Percepts: Video, sonar, speedometer, odometer, engine sensors, keyboard input, microphone, GPS
• Actions: Steer, accelerate, brake, horn, speak/display.
• Goals: Maintain safety, reach destination, maximize profits (fuel, tire wear), obey laws, provide passenger comfort.
• Environment: Urban streets, freeways, traffic, pedestrians, weather, customers.

The agent function is mathematical concept that maps percept sequence to actions.

f : P* → A

The agent function will internally be represented by the agent program.

The agent program is concrete implementation of agent function it runs on the physical architecture to produce f.

The vacuum-cleaner world: Example of Agent

• Environment: square A and B
• Percepts: [location and content] E.g. [A, Dirty]
• Actions: left, right, suck, and no-op