The Role of Agents in Simulation Models

Agent, Softbot, Simulation Model

An agent is any object that perceives its environment through sensors and acts upon that environment through effectors. Agents may be human agents, mechanical agents (robots) or software agents (softbots). It is assumed that each agent is endowed with a certain degree of intelligence. It means that the agents actions are controlled by his motives and action choices depend on his own experience rather than on built-in a priory knowledge. From this point of view, an agent is an autonomous object. The three basic motives of intelligent agents are drives, emotions and acts of will. Drives serve to satisfy the needs of an agent. These needs are e.g. food (energy), sexuality and safety at the lowest level, recognition, affection and sociability at the middle level and intellectual needs (an urge to find a meaning, striving to explain the world etc.) at the highest level. Elementary emotions are fear, joy, sadness, envy, anger, love, jealousy, shame, remorse, excitement, etc. Finally, acts of will are decisions, e.g. to buy a car, clean the room, go to school, etc. The role of software agents in simulation models depends on their intelligence and corresponding behaviour. From the opposite point of view, each software agent can be seen as a special model of human being that enables to simulate human behaviour at different levels of abstraction. The paper deals with agent behaviour, agent models, agent environments and with the role of agents in heterogeneous simulation models.

An agent is any object that perceives surrounding environment and acts upon that environment. Agents may be human agents, mechanical agents (robots) or software agents (softbots), etc. The paper deals with a possible role of agents in heterogeneous simulation models.