Particle Filter Implementation of PHD Filter

In this research, we implement a particle filter using the Probability Hypothesis Density (PHD) filter framework. The PHD filter is a random finite set-based approach that effectively handles nonlinear and non-Gaussian estimation problems. Our implementation includes key components such as particle initialization, weight update, and resampling algorithms. Additionally, we employ the Bearings-Only Tracking (BOT) measurement model, which provides accurate measurements of target bearing and range. The code implementation features measurement likelihood calculation and state update functions that process angular and distance observations. The BOT model integration involves coordinate transformation functions and measurement validation routines. This method enables more accurate tracking and prediction of target positions and movements, providing enhanced data precision for relevant applications. The implementation includes multi-target state extraction algorithms and clutter handling mechanisms, making it suitable for complex tracking scenarios. The particle filter structure maintains multiple hypotheses about target states, with systematic propagation and update cycles that improve estimation robustness.