Recursive Reductions of Action Dependencies for Coordination-Based Multiagent Planning

Abstract

Currently the most efficient distributed multiagent planning scheme for deterministic models is based on coordination of local agents’ plans. In such a scheme, behavior of other agents is modeled using projections of their actions stripped of all private information. The planning scheme does not require any additional information, however using such can be beneficial for planning efficiency. Dependencies among the projected public actions caused by sequences of local private actions represent one particular type of such information.

In this work, we formally define several types of internal dependencies of multiagent planning problems and provide an algorithmic approach how to extract the internally dependent actions during multiagent planning. We show how to take an advantage of the computed dependencies by means of reducing the multiagent planning problems and analyze worst-case privacy leakage caused by the used dependencies. We integrate the reduction method into a distributed multiagent planner and summarize other efficiency improving techniques used in the planner. We experimentally show strong reduction of majority of standard multiagent benchmarks and nearly doubling of solved problems in comparison to a variant of a planner without the reductions. The efficiency of the method is demonstrated by winning in a recent competition of distributed multiagent planners.