Building a Probabilistic Occupancy Map for Space Situational Awareness
Islam I. Hussein, Yue Wang, Richard Scott Erwin · 2011
In space situational awareness (SSA) a sensor network generally performs one of two main tasks: (1) search for new objects and (2) track detected objects. The main goal of the search task is to build a map of where objects are based on evidence collected by a sensor network. In a search mode, a sensor scans a sub-volume of the state-space with the objective of detecting new objects. Hence, the search space is partitioned into discrete cells that the sensors are schedule to scan. The challenge is that, when a measurement is made at a partition, transforming the sub-volume back to a sub-volume at an epoch time under Keplarian motion will result in the distortion of the geometric shape of the partition. In this paper a Bayesian methodology that seeks to reconstruct the object occupancy grid map at the epoch time given measurements made by the sensor network over time is developed. A simple planar scenario will be used to test the validity of the Bayesian map-building procedure and will be used to study the effect of sensor precision on the information content of the reconstructed object occupancy map. We conclude the paper with a discussion on the application of the methodology to SSA.