Autonomous Agent

autonomous agent

An entity that, by sensing and acting upon its environment, tries to fulfill a set of goals in a complex, dynamic environment. Properties:

1. it can sense the environment through its sensors and act on the environment through its actuators
2. it has an internal information processing and decision making capability
3. it can anticipate future states and possibilities, based on internal models (which are often incomplete and/or incorrect);
4. this anticipatory ability often significantly alters the aggregate behavior of the system of which an agent is part. An agent's goals can take on diverse forms:
   • desired local states
   • desired end goals
   • selective rewards to be maximized
   • internal needs (or motivations) that need to be kept within desired bounds. Since a major component of an agent's environment consists of other agents, agents spend a great deal of their time adapting to the adaptation patterns of other agents.

---

Thus the importance of MultiAgentSystems and the EmergentBehavior they tend to exhibit in the study of ComplexSystems. Nearly every system that a practitioner would acknowledge as "complex" is in some sense a MultiAgentSystem. --BillTozier

---

autonomous agent, more

When an agent has a certain independence from external control, it is considered autonomous. Autonomy is best characterized in degrees, rather than simply being present or not. To some extent, agents can operate without direct external invocation or intervention. Without any autonomy, an agent would no longer be a dynamic entity, but rather a passive object such as a part in a bin or a record in a relational table. Therefore, autonomy is considered by FIPA and the OMG's AgentsWorkingGroup [1] to be a required property of agents.

Autonomy has two independent aspects: dynamic autonomy and unpredictable autonomy. Agents are dynamic because they can exercise some degree of activity. An agent can have some degree of activity—ranging from simply passive to entirely proactive. For example, while ants are basically reactive, they do exhibit a small degree of proactivity when they choose to walk, rest, or eat. A supply-chain agent can react to an order being placed, yet be proactive about keeping its list of suppliers up to date.

Agents can react not only to specific method invocations but to observable events within the environment, as well. Proactive agents will actually poll the environment for events and other messages to determine what action they should take. (To compound this, many agents can be engaged in multiple parallel interactions with many other agents—magnifying the dynamic nature of the agent system.) In short, an agent can decide when to say "go."

Agents may also employ some degree of unpredictable (or nondeterministic) behavior. When observed from the environment, an agent can range from being totally predictable to completely unpredictable. For example, an ant that is wandering around looking for food can appear to be taking a random walk. However, once pheromones or food are detected, its behavior becomes quite predictable. In contrast, the behavior of a shopping agent might be highly unpredictable. Sent out to choose, negotiate, and buy a birthday present for your mother-in-law, the agent might return with something odd indeed or with nothing at all. In other words, the agent can also say "no."

--JamesOdell

---

See: WhatIsAnAgent

---

CategoryAgentOriented