The Canadian Space Agency has developed a multi-mission automated Collision Risk Assessment and Mitigation System (CRAMS). This paper describes the system and the challenges associated with its development and operation. The system receives e-mails and Conjunction Summary Messages (CSMs) from JSpOC, processes them and generates warning messages to the control centre when action is required based on predetermined thresholds. The system employs an approximate analytic probability model, and a conjunction geometry dependent hard body radius (HBR) for the primary object. The results of processing flight test data show accuracy extremely close to that of numerical integration in case of exact frame transformation. The limits of the simplifying assumptions for some transformations were also tested. To converge towards entry and exit (action -Stop action) criteria analysis was performed on past flight data and other data. The last challenge is the criterion on the quality of data given the fact that CSA receives only one or two data points before the time of conjunction. That implies that lack of sufficient data to test ability of covariance to predict miss distance variations for good quality data. One criterion was used for Radarsat-1 and Scisat based on not acting on any data of equivalent quality to that of TLEs. There has been previous decision not to act on TLE data after a visit and discussions with JSpOC.

No abstract

No abstract