Hide menu

The PARADOCS Project

PARADOCS in control of a simulated UAV logistics mission.

PARADOCS is a Prolog implementation of a reasoning system for a subset of TAL, that views Planning And Reasoning As DeductiOn with ConstraintS. It supports prediction from a fully instantiated set of actions, planning from the empty set of actions, and anything in between.

PARADOCS can be used to reason about a subset of TAL narratives that can be encoded using Horn clauses that are in close correspondence with the first-order TAL axioms. Temporal relations are managed by a general temporal constraint network implemented by a set of constraint handling rules (CHR), resulting in a plan structure that is more expressive than that of partial-order plans.

A plan generated by PARADOCS and its associated temporal constraint network.

One linearized instance of a plan generated by PARADOCS.

PARADOCS has been succeeded by the Logical Agents architecture.

Publications

[4] Martin Magnusson. Deductive Planning and Composite Actions in Temporal Action Logic. Licentiate thesis, Linköping University, Department of Computer and Information Science, September 2007. [ E-Press | .pdf ]
[3] Martin Magnusson and Patrick Doherty. Deductive Planning with Temporal Constraints. In Eyal Amir, Vladimir Lifschitz, and Rob Miller, editors, Logical Formalizations of Commonsense Reasoning: Papers from the 2007 AAAI Spring Symposium, Stanford, California, March 2007. AAAI Press. [ Conference | .pdf ]
[2] Martin Magnusson and Patrick Doherty. Deductive Planning with Temporal Constraints using TAL. In Xiaoping Chen, Wei Liu, and Mary-Anne Williams, editors, Proceedings of the International Symposium on Practical Cognitive Agents and Robots (PCAR), pages 141-152. University of Western Australia Press, November 2006. [ Conference | .pdf ]
[1] Martin Magnusson. Natural Language Understanding using Temporal Action Logic. In Proceedings of the Workshop on Knowledge and Reasoning for Language Processing KRAQ'06 at EACL'06, pages 29-36, April 2006. [ Conference | .pdf ]

Page responsible: Patrick Doherty
Last updated: 2014-04-30