IKT-BT: Indirect Knowledge Transfer Behavior Tree Framework for Multi-Robot Systems Through Communication Eavesdropping

IKT-BT: Indirect Knowledge Transfer Behavior Tree Framework for Multi-Robot Systems Through Communication Eavesdropping

1Cognitive Science Department, Rensselaer Polytechnic Institute    2College of Engineering, University of Georgia    3School of Computing, University of Georgia
IEEE Transactions on Cybernetics, 2025
@ARTICLE{10976677, author={Venkata, Sanjay Sarma Oruganti and Parasuraman, Ramviyas and Pidaparti, Ramana}, journal={IEEE Transactions on Cybernetics}, title={IKT-BT: Indirect Knowledge Transfer Behavior Tree Framework for Multirobot Systems Through Communication Eavesdropping}, year={2025}, volume={55}, number={6}, pages={2558-2570}, keywords={Robots;Eavesdropping;Knowledge transfer;Memory management;Scalability;Robot kinematics;Planning;Ontologies;Translation;Training;Behavior trees (BT);collective intelligence;eavesdropping;knowledge transfer;multirobot systems (MRS)}, doi={10.1109/TCYB.2025.3560564} }

This work explores an alternative approach inspired by eavesdropping mechanisms in nature that involves casual observation of agent interactions to enhance decentralized knowledge dissemination. We achieve this through a novel IKT-BT framework tailored for a behavior-based MRS, encapsulating knowledge and control actions in Behavior Trees (BT).

We present two new BT-based modalities - eavesdrop-update (EU) and eavesdrop-buffer-update (EBU) - incorporating unique eavesdropping strategies and efficient episodic memory management suited for resource-limited swarm robots. We theoretically analyze the IKT-BT framework for an MRS and validate the performance of the proposed modalities through extensive simulations in a multi-robot search and rescue scenario.

The framework demonstrates improved knowledge dissemination and collective performance compared to traditional direct communication approaches, while maintaining scalability and robustness in heterogeneous multi-robot systems.

IKTBT Overview
An overview of the IKTBT framework, showing indirect knowledge transfer through eavesdropping between agents.

Contributors

  • Sanjay Oruganti -- Cognitive Science Department, Rensselaer Polytechnic Institute (GitHub)
  • Ramviyas Parasuraman -- HeRoLab, School of Computing, University of Georgia (HeRoLab)
  • Ramana Pidaparti -- DICE Lab, College of Engineering, University of Georgia (DICE Lab)