Improving 3D food printing performance using computer vision and feedforward nozzle motion control

Abstract

3D food printing is an emerging technology to customize food designs and produce personalized foods. Food printing materials are diverse in rheological properties, which makes reliable extrusion-based 3D printing with constant printing parameters a challenge. Food printing often suffers from improper extrusion because of the varying elasticity of the food materials. In this study, a computer vision (CV)-based method is developed to measure the instant extrusion rate and width under constant extrusion pressure/force. The measured extrusion rate and extruded filament width were used to conduct a feedforward control of nozzle motion for a pneumatic 3D food printer. As a result, the CV-based control method improves extrusion line accuracy to 97.6–100% and prevents under-extrusion of white chocolate spread, cookie dough, and processed cheese. The method can also be used to customize filament width with less than 8% of deviation from the target. With a simple measurement setup and a user-friendly software interface, this CV-based method is deployable to most food printing applications to reduce trial-and-error experiments when printing a new food material.