Adaptive Constraint-Based Agents in Artificial Environments

[PARTIAL KNOWLEDGE]   [Unordered Domains]   [Ordered Domains]   [Probabilities]

[ Please note: The project has been discontinued as of May 31, 2005 and is superseded by the projects of the ii Labs. There won't be further updates to these pages. ]


(Related publications: [PUBLink] [PUBLink])

Probabilities for prospective states of the state resources are not a basic part of the model, because this is not generally needed and would waste a lot of system resources. In some cases, however, the use of probabilities can dramatically affect the plan result, especially if a state probability can be changed by special actions. For example, if the agent is searching for a key, it is not easy to express progress within the search process without probabilities. Whether the agent searches one drawer or two drawers must make a difference.

The probabilities can be realized by continuous domain state resources. The figure below shows a possible modeling of the search for the key.

[PARTIAL KNOWLEDGE]   [Unordered Domains]   [Ordered Domains]   [Probabilities]

For questions, comments or suggestions, please contact us.

Last update:
May 20, 2001 by Alexander Nareyek