Date of Award

8-2025

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Mechanical Engineering

Committee Chair/Advisor

Satchit Ramnath

Committee Member

Gregory Mocko

Committee Member

Cameron Turner

Abstract

The conceptual design stage of Unmanned Aerial Vehicles (UAVs) is an exploratory phase where designers are tasked with developing and evaluating a wide range of viable design concepts, considering different geometries and configurations based on mission requirements. The goal is to identify the most suitable design after the exploration for further development and progression to the next phase. Several tools and frameworks have been developed to assist designers in this stage, each offering unique capabilities, but all with the common aim of aiding in the efficient development and evaluation of conceptual designs. This is important in the UAV design process since the conceptual stage is the stage where changes can easily be applied without any substantial cost. However, research on existing frameworks which rely on the mission inputs from designers has revealed that many still limit designers in terms of freely exploring geometric designs and configurations. The objective of this research is to develop a mission-driven framework that enables designers to fully explore the conceptual stage, allowing them to generate and evaluate optimal designs without constraints on geometry or configuration. To achieve this, data from historical UAVs was collected, including information from literature on UAV designs, aircraft regulatory standards, and stored in a comprehensive database. Some of the data in the database were used to create models capable of predicting the geometric parameters required for UAV design. The remaining geometric parameters were either directly obtained or calculated from well-established models in literature, assisting in the design of the UAV’s geometric components. Once the geometric parameters were determined, a geometric modeling tool integrated into the framework is used to generate the UAV designs. Additionally, other parts of the collected data were utilized to build performance prediction models to evaluate the designed UAVs. To identify the best-performing designs, an optimization algorithm was incorporated into the frame- work, allowing it to determine the optimal designs generated during the conceptual stage. All these ii models and data are interconnected within the framework, with the designer’s mission input as the starting point. For this framework, only seven parameters are required from the designer, enabling the framework to explore the conceptual stage and provide a set of design solutions. A test run was conducted on the framework using seven mission parameters from an existing UAV. Using these parameters, the framework successfully generated and optimized designs with varying geometries and configurations, allowing for an in-depth exploration of the conceptual design stage. This was made possible through the seamless integration of models and an optimization algorithm within the framework, ensuring that the framework could freely and thoroughly explore the design space. The integrated models within the framework include experiential, empirical and analytical models, which enable the framework to leverage the designer’s mission inputs to estimate parameters, generate designs, and evaluate their performance. The trends became more pronounced in the final generation, once the solutions had converged. The test runs revealed that: • Larger UAV designs generally demonstrated improved aerodynamic efficiency. This happens when the increase in size resulted from an increase in the wing area. • Among designs of similar size, variations in configuration contributed to enhanced aerodynamic performance. This research presents a framework that enables designers to use mission requirements to navigate the conceptual design space, generating conceptual designs without constraints on geome- try and configuration. This is achieved through the integration of various models that automate the design stage by estimating the necessary parameters. Performance models are also included to test these designs. These models are seamlessly connected within the framework, ensuring automation and ease of use for designers. Additionally, the optimization algorithm integrated into the frame- work ensures the exploration of optimal designs during the conceptual stage, presenting designers with the best available options. This approach empowers the designer to make informed decisions while exploring the design space. Although the framework developed in this research is capable of generating new conceptual designs, the methods implemented do have their limitations. Current limitations and future improvements have been identified and discussed in detail in chapter 7.

Download

Available for download on Monday, August 31, 2026

Share

COinS