A Visit-Count-Based Complete Foraging Strategy for Robot Swarms

Author

• Arturo Yahir Gonzalez, The University of Texas Rio Grande Valley

Date of Award

5-1-2026

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Qi Lu

Second Advisor

Yifeng Gao

Third Advisor

Bin Fu

Abstract

Swarm robotics systems often rely on a balance between exploration and exploitation to perform tasks like Central Place Foraging. While exploitation methods such as pheromone trails and site fidelity are well-studied, the efficiency of the underlying random search exploration remains a challenge, frequently leading to redundant coverage and wasted time as multiple agents repeatedly search the same fruitless areas. This thesis introduces a memory-enhanced exploration strategy designed to improve the efficiency of random search in a swarm of simple robots.

In our proposed algorithm, each robot, constrained with limited memory, periodically logs its recent locations while exploring. This spatial data is shared at a central nest, which aggregates the information into a collective "visit map" representing the swarm's exploration history. This map is then used to intelligently guide future exploration; robots probabilistically avoid choosing destinations in areas with high visit counts, thereby discouraging redundant searches. The primary goal is to evaluate whether this collective memory system enables the swarm to find and collect all resources, particularly the final few clusters, more rapidly than a baseline swarm employing a purely random search. By comparing the performance of both strategies, this research aims to demonstrate that even with delayed, centralized information and minimal agent memory, a shared understanding of explored territory can significantly reduce task completion time and improve the overall efficiency of swarm foraging. Extensive simulation experiments demonstrate that the proposed strategy consistently outperforms the centrally placed baseline foraging algorithm (CPFA). Compared to CPFA, the proposed method reduces the total collection time by up to 33% and improves collection efficiency by more than 48% during the final stage of the mission.

Comments

Copyright 2026 Arturo Gonzalez. All Rights Reserved. https://proquest.com/docview/3371274618

Recommended Citation

Gonzalez, A. Y. (2026). A Visit-Count-Based Complete Foraging Strategy for Robot Swarms [Master's thesis, The University of Texas Rio Grande Valley]. ScholarWorks @ UTRGV. https://scholarworks.utrgv.edu/etd/1928