Ralph: Multiple Execution Architectures

Description

• An agent is defined to be a function f : P* --> D, where P* is the set of all perceptions available from the environment and D is the set of possible action decisions, either internal or external
• The method of choosing the best action depends on the type of knowledge available
• The multiple knowledge types define four different paths from perception to action, called Execution Architectures:
  • Decision-Theoretic
  • Goal-Based
  • Action-Utility
  • Condition-Action
• Each Execution Architecture runs in parallel, and the arbitrator chooses which action to take, based on environmental constraints