Abstract
In recent years, unmanned aerial vehicle (UAV) applications have been widely used in various manufacturing areas for the purpose of material handling or monitoring tasks. This situation increased the importance of proper estimation of UAVs' location. This paper presents hardware based Kalman Filter implementation for UAVs to accurately locate/detect its positions. To maintain high performance and compact form factor, Field-programmable Gate Array (FPGA) has been used as a hardware source. However, Kalman Filter algorithm needs lots of matrix computation and the typical implementation of matrix computations in hardware is complex and requires more effort than traditional software-based approaches. Matrix inversion computation in the Kalman gain formula is one of the most difficult matrix calculations in Kalman Filter algorithm and Chebyshev type inversion is used as a matrix inversion method to simplify hardware implementation. The proposed method simulated on both Matlab and Vivado based on the same scenario and numerical results of Kalman Filter and Chebyshev algorithm compared between these two simulation platforms. According to experimental results, the proposed solution serves compact and high performance standalone solution via FPGA for Kalman Filter implementation for UAVs.