Back to search

Marshall University

Very Large Scale Robotics path planning with Centroidal Voronoi Tessellation

Abstract

dc:description.abstract

Swarm robotics, also referred to as very large-scale robotics (VLSR), has emerged as a transformative approach for addressing complex tasks that are infeasible for single-robot systems. Applications range from environmental monitoring and disaster response to large-scale agricultural and industrial operations. However, as the number of robots in a swarm increases, so do the challenges associated with motion control, energy efficiency, and scalability. These challenges necessitate innovative solutions that balance microscopic robot behaviors with macroscopic system-level objectives. In this thesis, we address these challenges by building upon existing research [40], which introduced novel methods for optimizing swarm robotics systems using macroscopic and microscopic approaches. Specifically, we focus on enabling microscopic robots to conform to a reference Gaussian mixture model (GMM) distribution observed at the macroscopic scale. By optimizing the macroscopic level, we achieve an optimal overall system performance. However, a critical limitation of existing methods is their reliance on the systematic and global generation of Gaussian components (GCs) within obstacle-free environments to construct GMM trajectories. This requirement often introduces computational inefficiencies and limits adaptability in dynamic or complex environments. To overcome these limitations, this thesis proposes a novel approach that leverages centroidal Voronoi tessellation (CVT) for the systematic generation of GCs. CVT provides a robust framework for partitioning space into regions (Voronoi cells) and optimizing the placement of GCs within these regions. This method not only ensures consistency and reliability in trajectory generation but also significantly improves performance by reducing computational overhead and enhancing adaptability to environmental constraints. Through extensive simulations and theoretical analysis, we demonstrate the effectiveness of our approach in achieving scalable and energy-efficient swarm robotics systems. The results highlight the potential of CVT-based methods to bridge the gap between microscopic robot behaviors and macroscopic system goals, paving the way for more practical and deployable swarm robotics solutions. This thesis contributes to the growing body of knowledge in swarm robotics by providing a scalable, efficient, and reliable framework for motion control and energy optimization in large-scale robotic systems.

Degree

thesis:*
Name thesis:degree_name
Electrical and Computer Engineering
Level thesis:degree_level
Thesis
Discipline thesis:degree_discipline
College of Engineering and Computer Sciences

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Gao, Xu (James)
Contributors dc:contributor
  • Dr. Haroon Malik
  • Dr. Pingping Zhu
  • Dr. Mohammed Ferdjallah

Subjects

dc:subject × 8

Identifiers

dc:identifier.*
Repository record dc:identifier
https://mds.marshall.edu/etd/1961

Chain of custody

source
Harvested from
Marshall University
Base URL
mds.marshall.edu/do/oai/
Last updated
2026-11-21
Source record
OAI-PMH GetRecord
citation

Gao, Xu (James). Very Large Scale Robotics path planning with Centroidal Voronoi Tessellation. Thesis thesis, https://mds.marshall.edu/etd/1961