Date of Award

2025

Document Type

Open Access Master's Thesis

Degree Name

Master of Science in Mechanical Engineering (MS)

Administrative Home Department

Department of Mechanical and Aerospace Engineering

Advisor 1

Jung Yun Bae

Committee Member 1

Wayne W. Weaver

Committee Member 2

Vinh T. Nguyen

Abstract

This thesis presents the development of path planning algorithms for the coordination of heterogeneous robotic systems while considering size constraints. The objective is to generate practical and efficient solutions for real-world applications. The use of heterogeneous collaborative robots is beneficial in many applications, such as transportation operations in warehouses or manufacturing environments, surveillance, and monitoring, and task allocation and path planning are critical techniques that need to be addressed to deploy in real-world applications. This research focuses on automating lavender harvesting, where robots with varying capabilities must collaboratively navigate complex field layouts to efficiently complete harvesting tasks.

The problem considers the robot size constraints on the edges while solving task allocation and path planning to minimize the job completion time. It addresses the gap in this field by introducing two novel heuristic approaches. The first approach leverages Large Language Models (LLMs) to tackle generalized problems. We optimize prompts with detailed descriptions of task attributes, constraints, and environmental conditions, initially providing feasible solutions for small-scale problems. The LLM learns user requirements through iterative prompt refinement and generates feasible routes for each robot while minimizing the objective function. The second approach employs a greedy strategy, iteratively improving solution quality from an initial feasible solution while satisfying all relevant constraints. While the LLM-based approach accommodates a broader spectrum of problems, it may produce infeasible solutions, requiring careful implementation. Conversely, the greedy heuristic approach offers stability and guaranteed feasibility by simultaneously incorporating all the given constraints.

Both approaches are implemented and tested in simulations across multiple scenarios, varying the number of tasks, robots, and constraints. The computational results demonstrate the potential of these methods for real-world applications, offering valuable insights into solving complex path planning problems in heterogeneous multi-robot systems.

Available for download on Wednesday, October 01, 2025

Share

COinS