An Efficient Self-Supervised Learning Framework for Swarm Robot Trajectory Analysis

Authors

• Brooklyn Berry, The University of Texas Rio Grande Valley
• Gaukhar Nurbek, The University of Texas Rio Grande ValleyFollow
• Juan Manuel Perez, The University of Texas Rio Grande Valley
• Richard Tapia, The University of Texas Rio Grande Valley
• Qi Lu, The University of Texas Rio Grande ValleyFollow
• Yifeng Gao, The University of Texas Rio Grande Valley

Document Type

Conference Proceeding

Publication Date

11-2025

Abstract

Swarm robotics leverages groups of autonomous robots to perform complex tasks collaboratively. Recently, there has been growing interest in the Self-Supervised Learning framework (SSL) for social robots, yet very little research has been done on designing a SSL framework for foraging swarm robots systems. In this paper, to address the two challenges above, we proposed a novel efficient self-supervised learning framework. Specifically, we proposed 1) a shared weight multi-instance based encoder-decoder structure for model pre-training; and 2) an embedding series compression strategy to reduce the space cost in inference stage. Experiments show our system can match the performance of standard SSL frameworks while being more efficient.

Comments

Copyright © 2025, IEEE

Recommended Citation

Berry, Brooklyn, Gaukhar Nurbek, Juan Manuel Perez, Richard Tapia, Qi Lu, and Yifeng Gao. "An Efficient Self-Supervised Learning Framework for Swarm Robot Trajectory Analysis." In 2025 IEEE International Conference on Data Mining Workshops (ICDMW), pp. 2185-2189. IEEE, 2025. https://doi.org/10.1109/ICDMW69685.2025.00266

Publication Title

2025 IEEE International Conference on Data Mining Workshops (ICDMW)

DOI

10.1109/ICDMW69685.2025.00266