Welcome to the homepage of

     OAKplan 


A Case-Based domain-independent planner



OAKplan
is an innovative case-based planning system  which is able to efficiently retrieve planning cases from plan libraries with more than ten thousands  elements, heuristically choose a suitable candidate (possibly the best one) and adapt it to provide a good quality solution plan similar to the one retrieved from the case base.
Given a planning problem we encode it as a compact graph structure, that we call Planning Encoding Graph, which gives us a detailed description of the topology of the planning problem. By using this graph representation, we examine an approximate retrieval procedure based on kernel functions that leads to effectively face the problem of matching planning instances, allowing to achieve extremely good performance in standard benchmark domains.
Overall, we show that OAKplan is competitive with state of the art plan generation systems in terms of number of problems solved, CPU time, plan difference values and plan quality when cases similar to the current planning problem are available into the plan library.

Download of OAKplan 1.0 (executable code)
Download of OAKplan 2.0 (executable code also for metric-temporal domains)
Download of case and test problems (Metric and Temporal-Metric domains)

Main papers related to OAKplan