Semester

Summer

Date of Graduation

2022

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

Yu Gu

Committee Co-Chair

Jason Gross

Committee Member

Jason Gross

Committee Member

Guilherme Pereira

Committee Member

Nicholas Szczecinski

Committee Member

Xi Yu

Abstract

This thesis presents a study of different engineering design methodologies and demonstrates their effectiveness and limitations in actual robot designs. Some of these methods were blended together with focus on providing an easily interpreted project design flow while implementing more bottom-up, or feedback, elements into the design methodology. Typically design methods are learned through experience, and design taught in academia aims to shape and formalize previous experience. Usually, inexperienced engineers are taught approaches resembling the Verein Deutscher Ingenieure (VDI) 2221 process. This method presented by the Association of German Engineers in 2006 is regarded as the general system design process. This introductory process is largely left open to interpretation, and it is often unclear when to implement feedback in the design process. The objective of this thesis is to investigate the roles of top-down and bottom-up processes, and how to integrate them in the robot design methodology.

The proposed approach utilizes several components from existing design methods. There are three main conditional loops within the proposed approach. The first loop focuses on defining the problem in a top-down manner through logical decomposition, defining technical requirements, researching solutions, and conducting a trade study. These four steps are done iteratively until reaching the bottom of the system, the most primitive components. This is followed by a modeling and analysis loop. This works from the bottom to the top of the design in preparation for manufacturing and validation. The final loop of the proposed approach focuses on validation and verification. The testing and manufacturing involved allows for alterations to the design to fulfill the original technical requirements. These three loops occur until a proof of concept is achieved. The proposed method is intended to be applied iteratively. The first pass of the method results in a proof of concept, while the second results in a preproduction prototype, and the third in a production model. This assembly of design elements provides a project flow that leaves little to be interpreted and is suitable for small design teams while still flexible enough to be applied to diverse robotics projects.

This thesis provides three case studies analyzing the application of the hybrid design approach mentioned above to robotic system development. The first study showcases a complicated system design with a small development team. The second case is of simpler construction with a smaller developer team. This simpler case better demonstrates the benefits of this hybrid approach in robotic system development due to the comparatively higher speed at which the system matures. The third case study shows how this same proposed approach can be applied to the design of a bottom-up controlled swarm. These case studies are for future designers to reference as examples of the hybrid design methodology in application, and what can happen when there is a lack of feedback in design. This proposed hybrid design method can encourage design practices in new engineers that translate better to industrial applications, and therefore encourage faster integration of new engineers into established design engineering practices.

Download

Share

COinS