Orbit Determination of Sunlight Illuminated Objects Detected by Overhead Platforms

This thesis, authored by Captain Richard P. Osedacz in June 1989 at the Air Force Institute of Technology, addresses the technical challenge of determining the orbital elements of 'fastwalkers'—uncatalogued objects detected by overhead platforms via reflected sunlight. The research was conducted under the sponsorship of the Foreign Technology Division, Flight Performance Division, to assist the North American Aerospace Defense Command (NORAD) in identifying these objects, which were causing sensor saturation and creating concerns about potential uncatalogued satellites or database errors. The author develops a technique using a Gauss orbit determination method to establish an initial target state estimate, which is then refined using a batch weighted least squares estimator. The study concludes that orbit determination for fastwalkers is impossible using a single data set due to range ambiguity, as the least observable eigenvector is aligned with the line of sight. However, the author demonstrates that by utilizing multiple collections of the same object over consecutive days, a successful orbit determination can be achieved. The thesis includes an analysis of three specific events, including one unique case where an object was tracked over four consecutive days, yielding favorable results. The author notes that the slant range calculation is highly sensitive to data corruption and that the Gauss method is prone to failure if the data contains biases. The document also discusses the use of intensity history profiles as a 'fingerprint' to ensure that the same object is being tracked across different collections. The research emphasizes that while single-pass data is insufficient for precise identification, multi-day tracking provides a viable path for determining the orbital parameters of these objects.