Efficiently Reasoning with Interval Constraints in Forward Search Planning
Authors: Amanda Coles, Andrew Coles, Moises Martinez, Emre Savas, Juan Manuel Delfa, Tomás de la Rosa, Yolanda E-Martín, Angel García-Olaya7562-7569
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section we evaluate our approach in two ways. We compare the performance of our native approach to the compilation and to the MTP planner across a range of PDDL benchmark instances with interval constraints added. We then compare both approaches to the default planner provided with APSI 3.3.2 on a Mars rover domain originally written in DDL for an ESA project and translated to PDDL using the methods described in this paper.Our results in Table 3, show that while the compilation has the expressivity to model temporal relations, it is not feasible for solving larger problems as it introduces a large number of additional actions. |
| Researcher Affiliation | Collaboration | Department of Informatics, King s College London firstname.lastname@kcl.ac.uk European Space Agency, Oxford, UK Juan.Delfa@esa.int Department of Computer Science, Universidad Carlos III de Madrid {trosa,yescuder,agolaya}@inf.uc3m.es |
| Pseudocode | No | The paper does not include structured pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. |
| Open Datasets | Yes | We compared the performance of our native approach, the compilation and MTP on the following PDDL domains. For MTP, these were translated to MPDDL. Concrete (new): ... Crewplanning (IPC 2008): ... Zeno Travel (IPC 2000): ... Cafe (Halsey 2004): ... |
| Dataset Splits | No | The paper describes using PDDL domains and problems for evaluation but does not specify training, validation, or test dataset splits in the conventional sense of machine learning datasets. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | Yes | We then compare both approaches to the default planner provided with APSI 3.3.2 on a Mars rover domain originally written in DDL for an ESA project and translated to PDDL using the methods described in this paper. |
| Experiment Setup | No | The paper describes the technical details of the proposed approach and its integration into OPTIC, but it does not specify concrete experimental setup details such as hyperparameter values, specific training configurations, or system-level settings for its experiments. |