Unscented Kalman Filter (UKF) Program for Bearings-Only Tracking System

This document presents a MATLAB-based implementation of the Unscented Kalman Filter (UKF) for bearings-only tracking systems. The program employs sigma point transformation to handle nonlinear estimation problems, making it particularly valuable for target tracking applications in domains such as unmanned aerial vehicles (UAVs), self-driving cars, and robotic systems. The UKF algorithm works by propagating carefully selected sigma points through the nonlinear system model to approximate the state distribution more accurately than extended Kalman filters. Key implemented components include: - Sigma point generation using the unscented transform - Nonlinear state prediction and measurement update steps - Covariance matrix propagation for uncertainty quantification This implementation enables precise estimation of target states (position, velocity) and error covariance matrices, providing improved tracking accuracy for bearing-only measurements where only angular information is available. The code structure includes modular functions for system initialization, time update, measurement update, and results visualization, allowing users to analyze target trajectory patterns and positional data effectively. The program serves as both a practical tool for real-world applications and an educational resource for understanding nonlinear filtering techniques.