A UKF-PF based Hybrid Estimation Scheme for Space Object Tracking

TL;DR

This paper introduces a hybrid UKF-PF filter for space object tracking that combines the strengths of Kalman and Monte Carlo methods to improve accuracy and speed in estimating orbital states.

Contribution

The paper presents a novel hybrid filtering scheme that seamlessly integrates UKF and PF for enhanced space object tracking accuracy and efficiency.

Findings

Hybrid filter performs well across different orbital parameters.

Performance depends on the number of measurements within sensor FOV.

The method offers fast and accurate state estimation regardless of uncertainty.

Abstract

In this paper, we present a UKF-PF based hybrid nonlinear filter for space object tracking. Estimating the state and its associated uncertainty, also known as filtering is paramount to the tracking process. The periodicity of the Keplerian orbits and the availability of accurate orbital perturbation models present special advantages in filter design. The proposed nonlinear filter employs an unscented Kalman filter (UKF) estimate the state of the system while measurements are available. In the absence of measurements, the state pdf is updated via a sequential Monte Carlo method. It is demonstrated that the hybrid filter offers fast and accurate performance regardless of orbital parameters used and the amount of uncertainty involved. The performance of the filter under is found to depend upon the number of measurements recorded when the object is within the field of view (FOV) of the sensors.