Note: This agent works with chunking, which compiles the actions in the substates into rules that apply at the top-state. Use the command "learn -e" to turn on chunking, run the agent once, init-soar then run the agent again to see how the learned productions eliminate the need to subgoal in the second run.
- Hierarchical task composition via subgoaling
- John Laird
- Soar 8,9