I'm a Ph.D. candidate at the University of South Carolina working on planning and learning in dynamic, multi-agent environments. My work spans opinion networks, multi-agent path finding for robots, and ontology-driven tools for making planning decisions explainable.
Advised by Prof. Biplav Srivastava and Prof. Vignesh Narayanan at the AI4Society lab.
Published at:
A hybrid decentralized planning framework for multi-agent path finding that minimizes inter-robot information sharing. Achieves 2–510× reduction in coordination overhead and tested on physical TurtleBot4 robots.
Generalized planning combined with graph convolutional networks to select optimal intervention nodes in social networks, reducing misinformation spread by 86% on real-world Cora graph (2000 nodes). Award-winning AAAI 2024 demo.
OWL 2 ontologies for annotating and querying planning decisions. Extends to multi-agent settings via maPO; OMEGA tool generates natural-language explanations for MAPF executions. 95.2% user preference in evaluations.
OMEGA: An Ontology-Driven Tool for Explaining Multi-Agent Path Finding
maPO: An Ontology for Multi-Agent Path Finding and Its Usage for Explaining Planner Behaviour
Building a Plan Ontology to Represent and Exploit Planning Knowledge and Its Applications
On Generalized Planning for Controlling Opinion Networks: Interpreting Human-AI Dialog States and Beliefs
Towards Effective Planning Strategies for Dynamic Opinion Networks
Expressive and Flexible Simulation of Information Spread Strategies in Social Networks Using Planning
University of South Carolina · AI4Society Lab
Conducting research in automated planning, multi-agent systems, and reinforcement learning. Developing novel approaches for opinion network intervention, multi-agent path finding, and ontology-driven explainability tools.
IIT Kharagpur – Samsung Research
Conducted analysis on multi-resident smart home sensor data using time, frequency-based segmentation. Developed unsupervised algorithms for activity classification.