Model-Based Kalman Filter Design on an FPGA

This paper presents a design and real-time implementation of Kalman Filter on hardware considering a model-based approach. The filter is designed with different data types in the MATLAB Simulink environment using the Model Composer Blocks Library. After automatic code generation with Model Composer, the designed filter is exported to Vivado Design Suite as Intellectual Property (IP). Verification of the design is carried out on Field Programmable Gate Array (FPGA). By comparing the reference MATLAB model and realization on hardware, the design is verified for data accuracy. Moreover, performance outcomes for real-time applications like timing and utilization analyses are examined.