Lessons Learned from Implemented Software Architectures for Physical Agents
Papers from the AAAI Spring Symposium
Henry Hexmoor & David Kortenkamp,Cochairs
Technical Report SS-95-02
223 pp., $35.00
ISBN 978-0-929280-85-1
[Add to Cart] [View Cart]
The goal of this symposium is to shed light into reasons for architectural decisions in building artificial agents. Many important questions affect architectural decisions. For this symposium, we ask questions like the following only with respect to architectural decisions.
- How should the agent arbitrate/coordinate/cooperate its behaviors and actions?
- How can human expertise be easily brought into an agent's decisions?
- How much internal representation of knowledge and skills is needed?
- How should the computational capabilities of an agent be divided, structured, and interconnected?
- What types of performance goals and metrics can realistically be used for agents operating in dynamic, uncertain, and even actively hostile environments?
- Why should we build agents that mimic anthropomorphic functionalities?
- What, if any, role can advanced simulation technology play in developing and verifying modules and/or systems?
- How can a given architecture support learning?