Introduction to Agent Technology

Let discuss how agent technology evolved starting from conventional computing. AI technology was born in 1956. AI technologies are actually very much inspired by the nature(ANN — model of brain, genes and chromosomes etc.).

AI also has no single definition for intelligence. Some definitions of AI organized into 4 categories, which known as four schools of thoughts(Acting humanly/ Acting rationally / Thinking humanly / Acting rationally). The modern approach to AI is considered as building machines to do right thing(Acting rationally).

In the future, The gap between man-machine will reduce and lead to man-machine coexistent. In order to enable these development among other AI technologies, Agent technology as done a breakthrough.

Motivation for Agent technology

When we notice the real world(specially the modern world) which we are find that is very dynamic, interconnected, distributed, uncertain, involves many entities. A system with such features are call complex systems.

In a complex system, [ 1 + 1 > 2 ]. It means, performances of two entities together are higher than addition of performances of individuals(Team work). Complex system has another main feature, that removing a component of a system is does not destroy the system, i.e. still survive; otherwise call a complicated system.

Example: A family is a complex system, because removing one organ, we can still survive. But, motor car is a complicated system, because removing one wheel is can’t survive.

AI has introduce agent technology ( multi agent technology ) to model any complex system. Complex system are generally can’t model by classical theories such as, ohms, parade, newton laws etc. In classical theories, we have that common assumption that, Environment and conditions does not change when we executing the theory. But It’s not true when in reality or in complex, real world problems.

Examples for Complex systems :

1. A family with 5 members, because health condition of one member might enclose the system.
2. Brain with thousand million neuron is a complex system, with so many features (Connectivity, distributed, uncertainty)
3. Social medias like face book, consist of billion of connections.
4. Logistics in military attacks. Because in a war scenario, thousand of groups in army, navy, air-force, military, vehicles, food, medicine, water, ambulance should be dispatched in the appropriate manner.
5. Curio services. Because they are receiving and distributing from different parts from whole world.
6. Dynamic supply chain and dynamic project management.

• When modelling a CS, what we are really going to do is to define each entity as agents and program their communication leading to solution.

What is complexity?

These are many aspects of complexity.

1. Uncertainty : When there is a uncertainty about entities and the connections, the system come rather unpredictable.
2. Emergent property : That pop up due to interaction within the system, so that pop up feature is not belongs to any of the entries before handled.
3. Achieving goals : There are more than one way to achieve the same goal in CS’s.
4. Butterfly effect, Non-linearity : Complex systems are highly interdependent. Therefore small change in one location can propagate into a global change. For a example z-score issue created opportunities for expansion of Sri Lankan education system.

Inspiration for Agent

In the nature, massive and complex systems are govern by operations or computing happening at the tiny partial level. such as atom, molecule, ant ,bees , cell etc.

E.g :

• Sun is generating huge amount of energy due to atomic level reaction , known as nuclear fusion, converts hydrogen atoms into helium.
• Complex bee colonies are constructed by small bees, consist of a single queen, thousand of male drones and thousand of male drones and female worker bees.
• Complex human body is gain energy generated through glucose molecules.

Because of that, any complex system should be able to model by programming the behavior of tiny entities.

Characters of an Agent

1. Act on behalf of a master : The agent works only for his master. As a example, A servant at home works for owner of the house, not for the neighbors.
2. Autonomous : The agent work with little intervention or even no intervention. Servant work with little intervention or not, otherwise if servant want to be instructor to servant all at the time, no point with such agent.
3. Consume less resources : Servant doesn’t demand for much facilities and survive with little resources. For computer terms, agent is small refer to use of little CPU and memory.
4. Create on request, execute and disappear : The servant activate when necessary do the job and disappear after the job without instruction.
5. Proactive : The servant may do the cleaning, watering the garden. when master has not at home. This happens without instructions of the master.
6. Reactive : Suppose that servant is reactive if servant on the call but master ringing the bell. Then servant disconnect the phone and attend the masters request.
7. In additionally, agent may have other features. Intelligent, knowledge, rationality, mobility, interactive etc.

Agents are entities whose presence is noticeable to the environment or others.

“ An agent is anything that perceives it’s environment through sensors and act upon the environment through actuators “ — Russell & Norving

Agent and Environment

Agents are special kind of program, that make and affect to the environment as compared with normal computer programs. By looking at the environment, we can decide, whether they are exist an agent.

E.g : if the bus agent has operated just now, There may not passengers at the bus stop.

Types of software agents

Agent can be classified in many ways.according to service offers. Agent could be private or public. for a example personal assistance are private agents and Facebook, amazon, eBay are public agents. Agent could be stationary or mobile also, depending on from where it operate. Mobility is required to identify the best resource location to perform it’s roles.

Applications of Agents

Virtually any real world system can be model as agent solution. For a example, We can define customer agents, supplier agents, distributing agents, transporting agents etc. in a business. And model the interaction among them.

An another example in military scenario, we can have army, navy, aircraft, ambulance, truck agents, logistic agents etc.

Furniture manufacture, building chairs may introduce leg, arm, back, seat, wheel agents etc.

When taking electronic components such as resistors, transistors, capacitors, conductors etc. We can define these items as agents and simulate construction of circuits.