Aerospace System Prognostics and Health Management Model and Telemetry Task Optimization

Aerospace telemetry data are obtained through wireless transmission between sensors on the spacecraft and ground stations. Outliers are common in such data, which creates challenges to analyze them and extract useful information. Meanwhile, spacecraft and ground station equipment rely on manual maintenance, which requires significant manpower and financial resources. Although ground stations can handle the measurement and control tasks of multiple types of satellites in orbit, how to plan and allocate the tasks remains a challenge. The objective of this project is to develop prognostic and health monitoring systems for spacecraft systems and ground stations to diagnose and predict faults based on multivariate sensor data. Maintenance costs can be reduced, and system safety and reliability can be improved. In addition, task planning algorithms will be developed for spacecraft control mission to optimize the match with ground stations and multiple types of satellites, towards better efficiency and lower cost.In this project, we will 1) design data processing algorithms based on dynamic dimensionality reduction method to obtain informative features, 2) develop statistical inference and visual analytic methods for key components monitoring, fault detection and prediction utilizing remote sensing systems, 3) design planning algorithms to optimally allocate spacecraft measurement and control tasks.