Generative design - reality gap characterization
Automatic shape generation of printed parts optimized against external/internal goals. Any goal or combination of goals (multi-objective) can be used as long as they can be measured or calculated. Examples: strength/weight, momentum, moment of inertia, resonance frequency, temperature conductivity/expansion, aesthetics, specific shape etc.
At the ROBIN group we are using generative design for automatic optimization of physical shapes for 3D printing and CNC machining. We have our own custom developed generators based on a new high voxel count FEA modeling technique running on the latest parallel CPU/GPU hardware.
In this project the student will explore and characterize the reality gap between our own FEA simulators and various parts and shapes printed on our in-house printers (cnc).
- Matlab 2019
- Creality Ender, Ultimaker, Fortus 250, Markforged carbon, Formlabs SLA, HP JetFusion 540, Datron NEO cnc