Performance of Microwave Data Fusion in the Presence of Small Satellite Navigation and Attitude Errors

In recent years, small spacecraft have matured significantly into valuable assets used by the Earth science community. Some of the engineering benefits to a small spacecraft platform include lower cost, quicker development schedules, and the ability to purchase Commercial Off-The-Shelf (COTS) components. One of the primary observational benefits to this platform is the ability to launch multiple spacecraft in order to achieve faster revisit times. This is critical when observing phenomena that are fast-evolving or occur on a global scale. Despite these benefits, the limited mass, volume, and power of the spacecraft bus often constrains the spatial and radiometric resolution of hosted payloads. One proposed method to overcome these limitations is to combine multiple views of the same scene to form a higher quality image, a process known as data fusion.

In the area of passive microwave remote sensing, current techniques have vastly improved the state of the art in the field of data fusion. These techniques make use of partially correlated measurements from either a single scanning (linear or conical) radiometer or multiple fixed receivers on a single spacecraft. However, these techniques often don’t address the possibility of obtaining measurements from multiple spacecraft, or make assumptions that don’t reflect the engineering constraints of the observing system. These constraints often take the form of knowledge and control accuracy of both navigation and attitude states. For small satellites, this is especially true as available mass, volume, and power limits the quality of sensors and actuators on-board the spacecraft bus. Errors, which are inherent to all sensors and actuators, are often larger for small satellite COTS components. Furthermore, the limitations of the mission (power cycles, thermal management, etc.) may result in sporadic updates to the attitude and navigation solutions, resulting in growing error bounds in between measurement updates. The accumulation of both inherent and propagated errors from sensors and actuators translate to image registration errors and could potentially limit the ability to perform data fusion using small satellites.

The objective of this research is to determine the feasibility and effectiveness of data fusion for passive microwave radiometric measurements using multiple small spacecraft in the presence of real-world navigation and attitude errors. In this presentation, errors typical of modern small satellite COTS sensors with a Technology Readiness Level 9 (TRL-9) are introduced into simulated datasets. These datasets are then used to evaluate the performance of established microwave sensor fusion algorithms. The goal is to quantify how much the quality of the final fused image suffers in the presence of these real-world errors. A systems engineering discussion on current capabilities and suggested future improvements to small satellites in the context of data fusion is also provided. The results are expected to help constrain system level navigation and attitude requirements necessary to successfully combine small satellite microwave data sets for future mission designs.