Cost Sensitive Reachability Heuristics for Handling State Uncertainty
This work addresses scalability and plan quality optimization in conditional planning for AI applications, representing an incremental improvement over existing methods.
The paper tackled the problem of scaling non-deterministic conditional planning while optimizing plan quality metrics, such as actions with nonuniform costs, by introducing a novel generalization of planning graph based heuristics, resulting in empirical benefits compared to state-of-the-art planners.
While POMDPs provide a general platform for non-deterministic conditional planning under a variety of quality metrics they have limited scalability. On the other hand, non-deterministic conditional planners scale very well, but many lack the ability to optimize plan quality metrics. We present a novel generalization of planning graph based heuristics that helps conditional planners both scale and generate high quality plans when using actions with nonuniform costs. We make empirical comparisons with two state of the art planners to show the benefit of our techniques.