Project Description
This project requires the skills of a motivated student with a CS background to develop a high-performance graphical user interface that allows users to interact with Computational Pattern-Producing Networks. Our lab uses CPPN representations to control the geometry of soft robotic actuators, and investigates the application of automated design methods to discover new, novel soft actuator designs and material distributions. The output of this project will be an interactive software tool that allows users to explore the creation and modification of CPPNs fluidly, while interfacing with existing mesh triangulation and cloud storage software packages. Interest in soft robotics, automated design, graph theory, mesh generation, and simulation are a plus but not required.
Special Requirements
This project involves developing a web page with a graphical user interface for a neural-network based geometry generator; the data resulting from user interactions with the web page will be stored in a database. We have example code for the network-based geometry generation which can be used as a basis, so the project is not starting from scratch, but a successful applicant will likely need to have experience with the following (or similar related tools) for web page GUI and backend development: JavaScript, RUBY, PostgreSQL, WebGL, GUI development (choose your environment), C++, etc.
Contact
- Robert Maccurdy (faculty)