The proposed research will investigate the development of an AI-powered framework that converts user requirements into 3D-printable assistive technology designs using OpenSCAD. Many people with disabilities require customised tools to support everyday activities, but existing assistive devices may not adequately accommodate individual needs. Custom solutions can be expensive, time consuming to design, and difficult to produce without specialised engineering knowledge.

This project aims to address these limitations by exploring how large language models (LLMs) can automate the design process using natural-language descriptions of user needs. The framework will translate user requirements into parametric OpenSCAD models that can be modified and manufactured using 3D printing. It will also investigate iterative design refinement, in which AI-generated designs are reviewed, checked for code validity, and improved through repeated feedback cycles.

The research will evaluate the feasibility and reliability of this approach by testing whether the framework can generate valid, customisable, and practically useful designs for simple assistive technology applications. Evaluation will consider successful code generation, geometric validity, design iterations, and the extent to which the resulting devices satisfy the original requirements. Where feasible, selected designs will be 3D-printed and assessed for basic functionality.

The overall goal is to determine whether AI-driven design automation can make personalised assistive technology more accessible, reduce the expertise required to create customised devices, and provide a foundation for future systems capable of generating a broader range of functional, user-centred designs.

School

Biomedical Engineering

Research Area

AI; assistive technology; BME

Suitable for recognition of Work Integrated Learning (industrial training)? 

No

This project is hosted by the School of Biomedical Engineering with access to the facilities within the Tyree Foundation Institute of Health Engineering. 

The research will evaluate the feasibility and reliability of this approach by testing whether the framework can generate valid, customisable, and practically useful designs for simple assistive technology applications. Evaluation will consider successful code generation, geometric validity, design iterations, and the extent to which the resulting devices satisfy the original requirements. Where feasible, selected designs will be 3D-printed and assessed for basic functionality.

Please check with supervisory team.