Plato: Generic Ontology-Based Representation of Semantic Planning Knowledge for Automatic PDDL Generation. Jilg, D., Schultheis, A., & Bergmann, R. 2026. ![Plato: Generic Ontology-Based Representation of Semantic Planning Knowledge for Automatic PDDL Generation [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper doi abstract bibtex Artificial Intelligence (AI) planning can support automated decision-making in domains such as manufacturing, robotics, logistics, and autonomous systems. However, practical use requires explicit planning models, which are often created and maintained directly in the Planning Domain Definition Language (PDDL). This makes planning knowledge difficult to model, maintain, reuse, evolve, and adapt to different planners. This paper presents Plato, a generic ontology schema for representing semantic planning knowledge in a way that makes it easier to model, maintain, and adapt to different planners. Plato models central planning concepts such as typed entities, predicates, and actions as semantic objects. It is designed as a maintainable source representation from which planner-specific PDDL artifacts can be generated automatically. The ontology is supported by the Knowledge Transformation Framework, which transforms Plato ontologies into PDDL domain and problem files or even other semantic data structures. The proof-of-concept evaluation uses four example ontologies covering classical planning, temporal planning, and numeric planning with automatic discretization. The generated PDDL artifacts and plans are validated with VAL, and the planning tasks are solved with off-the-shelf planners. The results provide initial evidence that Plato can serve as a semantic source model for automatic planning-domain generation.
@conference{keod26_plato,
author={David Jilg and Alexander Schultheis and Ralph Bergmann},
title={Plato: Generic Ontology-Based Representation of Semantic Planning Knowledge for Automatic PDDL Generation},
booktitle={Proceedings of the 18th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KEOD},
year={2026},
pages={1021-1028},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0015351500004985},
isbn={978-989-758-868-6},
issn={2184-3228},
url = {https://www.wi2.uni-trier.de/shared/publications/KEOD_2026_Plato.pdf},
abstract = {Artificial Intelligence (AI) planning can support automated decision-making in domains such as manufacturing, robotics, logistics, and autonomous systems. However, practical use requires explicit planning models, which are often created and maintained directly in the Planning Domain Definition Language (PDDL). This makes planning knowledge difficult to model, maintain, reuse, evolve, and adapt to different planners. This paper presents Plato, a generic ontology schema for representing semantic planning knowledge in a way that makes it easier to model, maintain, and adapt to different planners. Plato models central planning concepts such as typed entities, predicates, and actions as semantic objects. It is designed as a maintainable source representation from which planner-specific PDDL artifacts can be generated automatically. The ontology is supported by the Knowledge Transformation Framework, which transforms Plato ontologies into PDDL domain and problem files or even other semantic data structures. The proof-of-concept evaluation uses four example ontologies covering classical planning, temporal planning, and numeric planning with automatic discretization. The generated PDDL artifacts and plans are validated with VAL, and the planning tasks are solved with off-the-shelf planners. The results provide initial evidence that Plato can serve as a semantic source model for automatic planning-domain generation.}
}
Downloads: 0
{"_id":"X47udf4ctwfgM9hpu","bibbaseid":"jilg-schultheis-bergmann-platogenericontologybasedrepresentationofsemanticplanningknowledgeforautomaticpddlgeneration-2026","author_short":["Jilg, D.","Schultheis, A.","Bergmann, R."],"bibdata":{"bibtype":"conference","type":"conference","author":[{"firstnames":["David"],"propositions":[],"lastnames":["Jilg"],"suffixes":[]},{"firstnames":["Alexander"],"propositions":[],"lastnames":["Schultheis"],"suffixes":[]},{"firstnames":["Ralph"],"propositions":[],"lastnames":["Bergmann"],"suffixes":[]}],"title":"Plato: Generic Ontology-Based Representation of Semantic Planning Knowledge for Automatic PDDL Generation","booktitle":"Proceedings of the 18th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KEOD","year":"2026","pages":"1021-1028","publisher":"SciTePress","organization":"INSTICC","doi":"10.5220/0015351500004985","isbn":"978-989-758-868-6","issn":"2184-3228","url":"https://www.wi2.uni-trier.de/shared/publications/KEOD_2026_Plato.pdf","abstract":"Artificial Intelligence (AI) planning can support automated decision-making in domains such as manufacturing, robotics, logistics, and autonomous systems. However, practical use requires explicit planning models, which are often created and maintained directly in the Planning Domain Definition Language (PDDL). This makes planning knowledge difficult to model, maintain, reuse, evolve, and adapt to different planners. This paper presents Plato, a generic ontology schema for representing semantic planning knowledge in a way that makes it easier to model, maintain, and adapt to different planners. Plato models central planning concepts such as typed entities, predicates, and actions as semantic objects. It is designed as a maintainable source representation from which planner-specific PDDL artifacts can be generated automatically. The ontology is supported by the Knowledge Transformation Framework, which transforms Plato ontologies into PDDL domain and problem files or even other semantic data structures. The proof-of-concept evaluation uses four example ontologies covering classical planning, temporal planning, and numeric planning with automatic discretization. The generated PDDL artifacts and plans are validated with VAL, and the planning tasks are solved with off-the-shelf planners. The results provide initial evidence that Plato can serve as a semantic source model for automatic planning-domain generation.","bibtex":"@conference{keod26_plato,\nauthor={David Jilg and Alexander Schultheis and Ralph Bergmann},\ntitle={Plato: Generic Ontology-Based Representation of Semantic Planning Knowledge for Automatic PDDL Generation},\nbooktitle={Proceedings of the 18th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KEOD},\nyear={2026},\npages={1021-1028},\npublisher={SciTePress},\norganization={INSTICC},\ndoi={10.5220/0015351500004985},\nisbn={978-989-758-868-6},\nissn={2184-3228},\nurl = {https://www.wi2.uni-trier.de/shared/publications/KEOD_2026_Plato.pdf},\nabstract = {Artificial Intelligence (AI) planning can support automated decision-making in domains such as manufacturing, robotics, logistics, and autonomous systems. However, practical use requires explicit planning models, which are often created and maintained directly in the Planning Domain Definition Language (PDDL). This makes planning knowledge difficult to model, maintain, reuse, evolve, and adapt to different planners. This paper presents Plato, a generic ontology schema for representing semantic planning knowledge in a way that makes it easier to model, maintain, and adapt to different planners. Plato models central planning concepts such as typed entities, predicates, and actions as semantic objects. It is designed as a maintainable source representation from which planner-specific PDDL artifacts can be generated automatically. The ontology is supported by the Knowledge Transformation Framework, which transforms Plato ontologies into PDDL domain and problem files or even other semantic data structures. The proof-of-concept evaluation uses four example ontologies covering classical planning, temporal planning, and numeric planning with automatic discretization. The generated PDDL artifacts and plans are validated with VAL, and the planning tasks are solved with off-the-shelf planners. The results provide initial evidence that Plato can serve as a semantic source model for automatic planning-domain generation.}\n}\n\n","author_short":["Jilg, D.","Schultheis, A.","Bergmann, R."],"key":"keod26_plato","id":"keod26_plato","bibbaseid":"jilg-schultheis-bergmann-platogenericontologybasedrepresentationofsemanticplanningknowledgeforautomaticpddlgeneration-2026","role":"author","urls":{"Paper":"https://www.wi2.uni-trier.de/shared/publications/KEOD_2026_Plato.pdf"},"metadata":{"authorlinks":{}}},"bibtype":"conference","biburl":"www.wi2.uni-trier.de/publications/WI2Publikationen.bib","dataSources":["MSp3DzP4ToPojqkFy","9v4v84gbRB6iyKeHb","FWPbCYSxS5WYzo747","YqH7TFm9yMJqzRPy9"],"keywords":[],"search_terms":["plato","generic","ontology","based","representation","semantic","planning","knowledge","automatic","pddl","generation","jilg","schultheis","bergmann"],"title":"Plato: Generic Ontology-Based Representation of Semantic Planning Knowledge for Automatic PDDL Generation","year":2026}