You are here: Objectives» Contributions» Multi-Agent Planning » Cooperative Planning

Cooperative Planning

The Cooperative Multi-Agent Planning System (MAP) is a general-purpose tool based on Partial Order Planning (POP) and suitable to cope with both tightly-coupled and loosely-coupled cooperative multi-agent planning problems.

Planning agents incorporate an embedded POP system, modified to compute partial solutions instead of complete plans. The resolution process starts with the construction of a distributed Relaxed Planning Graph, which allows agents to exchange planning information. Agents coordinate by performing a refinement planning process by which they incrementally refine an initially empty plan until a solution plan is obtained.

Planning tasks are described in a PDDL3.1-based MAP language with some extensions to support privacy and partial information. The user should provide a domain and problem file per agent, describing the information managed by the agent.

Currently, there are two different versions of the Cooperative Multi-Agent Planning System in development:

  • MAP-POP: the initial approach to Cooperative MAP uses a backwards-chaining POP and guides the plan-space search through standard, state-of-the-art POP heuristic functions.
  • FMAP: this version of the Cooperative MAP System features a forward-chaining POP and a custom heuristic function based on Domain Transition Graphs. FMAP is fully integrated in the PlanInteraction architecture.
Relevant papers:

Alejandro Torreño, Eva Onaindia, Óscar Sapena
FMAP: Distributed Cooperative Multi-Agent Planning
Applied Intelligence, (In Press), (2014)

Alejandro Torreño, Eva Onaindia, Óscar Sapena
A Flexible Coupling Approach to Multi-Agent Planning under Incomplete Information
Knowledge and Information Systems, Vol. 38(1), pp. 141-178, (2014)

Alejandro Torreño, Eva Onaindia, Óscar Sapena
FMAP: A Heuristic Approach to Cooperative Multi-Agent Planning
Preprints of the ICAPS'13 DMAP Workshop on Distributed and Multi-Agent Planning, pp. 84-92, (2013)

Alejandro Torreño, Eva Onaindia, Óscar Sapena
An Approach to Multi-Agent Planning with Incomplete Information
20th European Conference on Artificial Intelligence (ECAI 2012), pp. 762-767, (2012)