Planning with Numeric Timed Initial Fluents

Authors: Chiara Piacentini, Maria Fox, Derek Long

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this paper we present an extension of the planner POPF2 (POPF-TIF) to handle problems with numeric Timed Initial Fluents. We propose and evaluate two contributions: the first is based on improvements of the heuristic evaluation, while the second considers alternative search algorithms based on a mixture of Enforced Hill Climbing and Best First Search. 4 Experimental Evaluation We now present results of the planner POPF-TIF with the following domains (30 problems each domain):
Researcher Affiliation Academia Chiara Piacentini, Maria Fox, Derek Long Informatics Department King s College London London, UK email: name.surname@kcl.ac.uk
Pseudocode No The paper describes algorithms but does not provide structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statement or link indicating the availability of open-source code for the methodology described.
Open Datasets Yes Unit Commitment Problem (UCP); Rover domain: a variant of the numeric rover domain in which the rover is equipped with a solar panel that gives energy to the rover according to the exposure to the sun. Temperature domain: a PDDL version of the problem presented in (Ono, Graybill, and Williams 2012), where the objective is to maintain the temperature of a smart home within the user s preferences. Skier domain:
Dataset Splits No The paper mentions '30 problems each domain' but does not provide specific dataset split information (e.g., percentages, sample counts, or explicit validation splits) needed for reproduction.
Hardware Specification Yes All tests used a 3.4 GHz Intel Core i7-2600 machine, limited to 30 minutes and 4 GB of memory.
Software Dependencies No The paper mentions planners like POPF2 and UPMurphi but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes For each search strategy we use heuristic evaluations with a different number of lookaheads, from 0 to 10, where 0 indicates the standard heuristic evaluation.