Inline Coherent Imaging Based Control of Laser Machining Processes using a Field-Programmable Gate Array

Abstract

Lasers are becoming an increasingly popular tool for use in industrial manufacturing settings. Their high power density, contact-free nature, and ease of automation give them an edge over traditional methods in processes such as welding, cutting, and drilling. However, laser beams are a difficult tool to control since they are made of light. Therefore, they cannot be monitored in the same way as a mechanical tool with constant physical dimensions. A recent invention, inline coherent imaging (ICI), uses the interference properties of light to monitor the depth of the processing beam at high speeds. However, the computational demands of ICI data processing are intensive, and it is difficult for standard computers to keep up with the high volumes of data that are output by an ICI system. As a result, the real-time control capabilities of ICI are limited by the rate at which the computer can process the data. In this work, I present a system that uses a field-programmable gate array (FPGA) to process ICI data with a high level of determinism and low latency compared to computer-based ICI. The algorithms that were developed to take advantage of the FPGA's architecture are presented, followed by a detailed description of how they were implemented programmatically. I also describe the hardware that was used to perform experiments with closed-loop ICI feedback control. Finally, I present the results of tests done with two applications, autofocus and welding depth control. I show that FPGA-based ICI achieves superior performance to computer-based control, pushing the latency of an ICI calculation down from 3600±900 μs to 54±5 μs. Areas of future work include fine-tuned welding control, implementation of ASIC and/or advanced interpolation or FFT algorithms, and investigation of other laser processes such as additive manufacturing and surgical bone ablation.