A Lightweight Cubature Kalman Filter for Attitude and Heading Reference Systems Using Simplified Prediction Equations

Researchers have unveiled the Kaisoku Cubature Kalman Filter (KCKF), a new lightweight algorithm that significantly speeds up orientation and motion sensing for robots and drones. It reduces computation time by approximately 19% on high-performance computers and 15% on low-cost hardware like single-board computers, while maintaining full accuracy. This is achieved by simplifying complex prediction equations, requiring fewer floating-point operations than the standard Cubature Kalman Filter it improves upon.