Case Study: FLIR Systems achieves up to 60% faster FPGA prototyping and rapid algorithm-to-hardware turnaround with MathWorks MATLAB & HDL Coder

Cutting Algorithm Development Time with MATLAB: Q&A with FLIR

FLIR Systems, a global designer and manufacturer of advanced sensing technologies, faced long hardware prototyping cycles because hardware engineers translated algorithm engineers’ designs into HDL from written specs. That workflow made it hard to know whether mismatches between FPGA implementations and simulations were caused by the algorithm or the HDL, and even small algorithm changes required extensive HDL rewrites. To address this, FLIR turned to MathWorks’ MATLAB and HDL Coder for model-based design.

Using MathWorks’ tools, FLIR generated synthesizable HDL directly from algorithm code, implemented and tested designs on FPGA, and verified results against simulations. The new workflow cut prototyping time by up to 60%, increased MATLAB code reuse for HDL generation from 0% to 30%, and allowed a major algorithm change that once took six weeks to be implemented in three hours — delivering working camera prototypes that matched simulations.

FLIR Systems

Nicholas Hogasten

Manager of Image Processing Technology