MATLAB Simulation Suite Featuring Square-Root Cubature Kalman Filter (CKF), Unscented Kalman Filter (UKF), and Extended Kalman Filter (EKF)

This paper presents a MATLAB simulation package that implements three advanced filtering algorithms: Square-Root Cubature Kalman Filter (CKF), Unscented Kalman Filter (UKF), and Extended Kalman Filter (EKF). The simulation programs are designed with modular architecture, allowing users to easily modify system models, measurement functions, and noise characteristics for specific control system applications. Each algorithm implementation includes detailed code comments explaining the mathematical foundations and implementation steps. The CKF implementation utilizes spherical-radial cubature rules for numerical integration, while the UKF employs the unscented transform with carefully chosen sigma points. The EKF implementation demonstrates first-order Taylor series linearization techniques for state estimation. The code structure features separate function files for prediction and update steps, with configurable parameters for process noise covariance (Q) and measurement noise covariance (R). Users can extend the package by adding additional filtering algorithms or modifying existing ones to meet specific research requirements. These simulation tools provide control system designers and researchers with powerful resources for algorithm optimization, performance comparison, and achieving improved system performance metrics through practical implementation examples.