As resident space object populations grow and satellite propulsion capabilities improve, it will become increasingly challenging for space-reliant nations to maintain space domain awareness using current human-in-the-loop methods. This work presents an adaptive approach to autonomous sensor network management for tracking multiple maneuvering and nonmaneuvering satellites with a diversely populated space object surveillance and identification network. The proposed method integrates a suboptimal partially observed Markov decision process (POMDP) with a static multiple-model adaptive unscented Kalman filter to task sensors and maintain viable orbit estimates for all targets. The POMDP uses largest Lyapunov exponent, Fisher information gain, and sensor transportability metrics to assess the reward for tasking a specific sensor to track a particular satellite. Like in real-world situations, the population of target satellites vastly outnumbers the available set of sensors. Without a robust and adaptable tasking algorithm, the sensor network would fail to track all targets. The strategy presented herein successfully tracks 203 nonmaneuvering and maneuvering spacecraft using only 24 ground- and space-based sensors. The results show that multiple-model adaptive estimation coupled with a multimetric, suboptimal POMDP can effectively task a diverse network of sensors to track multiple maneuvering spacecraft, while simultaneously monitoring a large number of nonmaneuvering objects.
Adaptively Tracking Maneuvering Spacecraft with a Globally Distributed, Diversely Populated Surveillance Network
Journal of Guidance, Control, and Dynamics ; 42 , 5 ; 1033-1048
2018-12-18
16 pages
Article (Journal)
Electronic Resource
English
British Library Conference Proceedings | 2017
|Adaptively robust unscented kalman filter for tracking a maneuvering vehicle
Online Contents | 2014
|Maneuvering Spacecraft Tracking via State-Dependent Adaptive Estimation
Online Contents | 2016
|