Improvements in Accuracy of Fused Deposition Modeling Via Integration of Low-Cost On-Board Vision Systems

Abstract

Abstract In fused deposition modeling (FDM) process, several uncertainty sources contribute to low accuracy and bad surface finishing. Besides intrinsic uncertainties, other factors, related to the machine structure and actuation, participate to increase complexity and reduce the controllability of the process. On-board vision systems can be successfully exploited to monitor the deposition process and to acquire quantitative information on geometry and transients. This information can be used to calibrate the process parameters where a compensation strategy is implemented. In this paper, a low-cost vision system capable of microscopic acquisitions is integrated into an FDM machine and image-processing techniques are exploited to define calibration procedures to improve the process accuracy, thus extending the use of this equipment to micromanufacturing applications. The effectiveness of the integration is evaluated through the fabrication of a microfluidic device.