Efficient Fuzzy Controllers for FPGA using High Level Synthesis

Fuzzy Logic Controllers (FLC) are control systems commonly used on problems where data is not accurate or its domain is not well-known. This is because instead of using complex mathematical models to work, they use a set of rules to evaluate data. To implement this kind of controllers one interesting option is FPGA. This technology has advantages based on reconfigurability, performance, energy usage and design flexibility. This work presents a tool based on HLS and FPGA that allows users to generate Fuzzy Logic Controllers from abstract descriptions. Also the most impactful synthesis directives for optimizing the different stages of a FLC are detailed. Finally, three case studies are presented to evaluate the tool.