Oversubscription Planning as Classical Planning with Multiple Cost Functions

Abstract

The aim of classical planning is to minimize the summed cost of operators among those plans that achieve a fixed set of goals. Oversubscription planning (OSP), on the other hand, seeks to maximize the utility of the set of facts achieved by a plan, while keeping the cost of the plan at or below some specified bound. Here, we investigate the use of reformulations that yield planning problems with two separate cost functions, but no utilities, for solving OSP tasks. Such reformulations have also been proposed in the context of netbenefit planning, where the planner tries to maximize the difference between the utility achieved and the cost of the plan. One of our reformulations is adapted directly from that setting, while the other is novel. In both cases, they allow for easy adaptation of existing classical planning heuristics to the OSP problem within a simple branch and bound search. We validate our approach using state of the art admissible heuristics in this framework, and report our results.

Selecting goals in oversubscription planning using relaxed plans

“…Probably due to the IPC, most work on PSP has been on net-benefit [13,19,18,20,21,22,23]. In contrast, OSP has been less explored [1,2,3,4,5], even though the existence of a limited resource (time, fuel, battery, storage space, money, . .…”

“…A recent work presents two compilations transforming the OSP problem into a multiple cost function (MCF) planning task [4], where utilities form the primary cost function and the costbound the secondary one. Their first reformulation, soft-goals, is a straightforward adaptation of Keyder and Geffner's one [27].…”

“…In its original formulation a limited consumable resource, the battery of a rover, makes reaching all goals impossible; there are more goals than those that can be achieved given its stored energy. Most papers in the literature generalize the problem by representing the insufficient resource as a cost-bound of the plan [1,2,3,4,5]. Only plans with cost equal or lower than the bound will be valid, with the assumption that no plan will exist reaching all goals with such cost bound.…”

“…OSP appears in many real applications but surprisingly it has not drawn too much attention from the planning community after the seminal work of Smith [1]. Only recently some works have addressed this task, dealing with its optimal version [2,3,4,5]. But the suboptimal OSP task remains mostly unexplored even if solving OSP tasks optimally is not feasible in many real cases.…”

Evaluating Plan-Property Dependencies: A Web-Based Platform and User Study

“…Fixed utility is a standard form of oversubscription planning, which could be solved optimally using known algorithms (e. g (Smith 2004;Domshlak and Mirkis 2015;Katz et al 2019)…”