Home  /  Projects  / 

Knowledge Engineering for Planning

Contact: Students: Bharath Muppasani, Vishal Pallagani, Ritirupa Dey; Faculty: Biplav Srivastava, Vignesh Narayanan

Key Idea

This research focuses on leveraging knowledge engineering, specifically through the development of ontologies, to enhance the capabilities of AI planning systems. By structuring domain and planner knowledge, we aim to improve planner selection, optimize performance through learned constraints (macros), and provide transparent, human-readable explanations for complex behaviors in domains like Multi-Agent Path Finding (MAPF).

Planning Ontology for Performance Improvement

We hypothesize that given a large number of available planners and diverse planning domains, they carry essential information that can be leveraged to identify suitable planners and improve their performance. We use data from the International Planning Competition (IPC) to construct a planning ontology and demonstrate that it can lead to the selection of promising planners and improve their performance using macros—a form of action ordering constraints extracted from the ontology.

Planning Ontology Tool

Available on Planning.Domains, this interactive tool transforms PDDL (domain, problem, plans) into RDF/OWL knowledge graphs. It features D3.js-based visualization and SPARQL querying capabilities to explore planning metadata.

Resources: [PURL] [Plugin Repository]

Planning Ontology Tool Demo

Representative Publications

  • Building a Planning Ontology to Represent and Exploit Planning Knowledge and Its Applications
    Bharath Chandra Muppasani, Nitin Gupta, Vishal Pallagani, Biplav Srivastava, Raghava Mutharaju, Michael N. Huhns & Vignesh Narayanan
    Discovery Data Journal, 3, 55 (2025). https://doi.org/10.1007/s44248-025-00093-9
    [Paper] [Website]
  • Building a Plan Ontology to Represent and Exploit Planning Knowledge and Its Applications
    Bharath Chandra Muppasani, Nitin Gupta, Vishal Pallagani, Biplav Srivastava, Raghava Mutharaju, Michael N. Huhns & Vignesh Narayanan
    Eighth International Conference on Data Science and Management of Data (CODS-COMAD '24), India, 2024.
    [Paper] [Website]

Representative Patents

  • USC 1649 Improving Planner Performance by Learning and Using Metadata of Experiences
    Biplav Srivastava, Vishal Pallagani, Bharath Muppasani
Planning Ontology Diagram

maPO: A Multi-Agent Planning Ontology for Explainable Path Finding

As multi-agent systems become more autonomous, the need for transparent and interpretable decision-making is critical. To address this, we introduce the Multi-Agent Planning Ontology (maPO), a unified semantic schema that formalizes MAPF concepts. Our framework includes a log-to-graph pipeline that ingests planner execution traces and transforms them into a rich knowledge graph. This allows stakeholders to ask complex questions about the planner's behavior and receive human-readable explanations.

Representative Publication

  • maPO: An Ontology for Multi-Agent Path Finding and Its Usage for Explaining Planner Behaviour
    [Paper] [Website]
  • OMEGA: An Ontology-Driven Tool for Explaining Multi-Agent Path Finding
    Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence - Demonstrations Track (AAAI-26 Demo), Singapore, January 2026 (AR=27%)
    Bharath Muppasani, Ritirupa Dey, Biplav Srivastava, Vignesh Narayanan
    [Paper] [Video] [Poster]
Planning Ontology Diagram