PEAS in Artificial Intelligence

Peas in Artificial Intelligence is the measuring unit used to measure efficiency of a system. PEAS also provides accountability for a system's design.

To understand PEAS better you can have idea from statement:

"An agent perceives its environment via sensors and act upon it through actuators."

PEAS in Artificial Intelligence is Measuring Unit for Testing Systems

An AI system is composed of an agent and its environment. In their setting, the agents act. The environment may contain other agents. An agent is anything that can be seen as:

Perceiving its environment by sensors.
Acting on it by actuators

Abbreviation of PEAS is:

P- Performance
E- Environment
A-Actuators
S-Sensors

Starting off with Performance in our discussion on peas in artificial intelligence.

Performance| PEAS in Artificial Intelligence:

"An objective criterion for success of an agent's behavior."

For optimum performance rational agent plays the most vital role. Performance in Artificial Intelligence can also be simply explained as, how much durable and well acting a system is in different precepts handling.

Take Example of a Bus Journey to Tokyo: The performance measure of this journey is Safety, Arrival to correct destination, On time departure and smooth driving.

Performance also varies with the type of agent acting in each precept. Each of these agents have different functions and differ in abilities. To name some of the main agents in artificial intelligence:

Table Look Up driven Agents
Simple Reflex Agents
Model-Based Reflex Agents
Goal-Based Agents
Learning Agents
Rational Agents
Utility-Based Agents

The Key agent in performance measure is rational agent, lets discuss little more about what rational agent is?

What is a Rational Agent in AI?

Rational Agents in Artificial Intelligence

A rational Agent, for each possible precept selects an action that maximizes performance measures. Rational Agents are logical workers and analyze all alternative paths and multiple ways to complete the task and go for the shortest and least cost way.

Rational agents are considered as smartest agents when it comes for decision making abilities.

Example for Rational Agent in Artificial Intelligence:

Rational Agent in our example for bus journey can be considered as a driver who drives in proper lanes. Who will depart and arrive on time and take care of passengers safety through out this journey.

Environment for the System in Artificial Intelligence:

The space or place in which system is kept or agents can act is known as environment. For example taking the most famous example of a Taxi Driver case, For a Taxi roads, blocks and Highways are its environment.

Environment for agents in AI are of 5 Major types. Types of environment are following.

Fully Observable/ Partially observable
Episodic/ Static Environment
Static/ Dynamic
Discrete/ continuous
Deterministic/ Stochastic

Actuator in Artificially Intelligent Systems:

Actuators are the components of System through which an agent acts upon its environment. Agent cannot perform any action if Actuators are not present.

Consider your daily life example:

You want to drive a car. Here in driving case, Roads are the environment. Now steering, accelerator and brakes are actuators. These actuators will help you to act.

Sensors in Artificially Intelligent Systems:

Sensors act as the eyes to every system. With help of Sensors, agents are able to sense the environment and understand what’s going on.

Take example of a bar code scanning system, which produces bill at counter. Until you let it scan the items you want, it wouldn't be able to bill it (No action taken without sensing). Thus, you won't be able to take the items home.

Summary:

PEAS in artificial intelligence stands for performance, Environment, Actuators and sensors.
Agents are those who act upon environment. They are the action makers.
Sensors are used by system agents to sense the environment and understand the environment in which they are.
With actuators actions take place.
Most well performing agents are rational agents.
Environment is of 5 main types.
There are 6 main types of agents.

HowTechyy

PEAS in Artificial Intelligence - Definitions and Examples

PEAS in Artificial Intelligence