1994 Workshop on Agent Theories, Architectures, and Languages (ATAL 94), August 1994, Amsterdam, Netherlands
This paper presents a new architecture for controlling autonomous agents in dynamic multi-agent worlds, building on previous work addressing reactive and deliberative control methods. The proposed multi-layered architecture allows a rationally bounded, goal-directed agent to reason predicatively about potential conflicts by constructing causal theories which explain other agents' observed behaviours and hypothesize their intentions; at the same time it enables the agent to operate autonomously and to react promptly to changes in its real-time environment. A principal aim of this research is to understand the role different functional capabilities play in constraining an agent's behaviour under varying environmental conditions. To this end, an experimental test bed has been constructed comprising a simulated multi-agent world in which a variety of agent configurations and behaviours have been investigated. A number of experimental findings are reported.
Proceedings of the 1994 Workshop on Agent Theories, Architectures, and Languages (ATAL 94).