Rev 24 5 Aug 2010 1 OPEN SOURCE SOFTWARE SUITE FOR SPACE SITUATIONAL AWARENESS AND SPACE OBJECT CATALOG WORK EUROPEAN SPACE ASTRONOMY CENTRE (ESA/ESAC), MADRID, SPAIN 3 – 6 MAY 2010 Paul J. Cefola (1) , Brian Weeden (2) , Creon Levit (3) (1) Consultant, 59 Harness Lane, Sudbury Massachusetts 01776, USA, (also Adjunct Faculty, University at Buffalo, SUNY), Email: paulcefo@buffalo.edu (2) Technical Advisor on Space Security and Sustainability Issues, Secure World Foundation, 5610 Place Bayard, Brossard, Quebec, J4Z 2A5, Canada, Email: bweeden@swfound.org (3) Chief Scientist for Projects and Programs, NASA Ames Research Center, Moffet Field, California 94035, USA, Email: creon.levit@nasa.gov ABSTRACT The accuracy of the orbital data products used for space situational awareness is affected by the evolution of the sensors collecting the data, the knowledge and control of the errors in the sensor network, the knowledge of the space environment, the available computing resources (both hardware and software), and the number of space objects to be monitored. While the number of objects in orbit has grown significantly over the last three decades, the quality and quantity of the orbital data products, particularly those available to non-government space operators has not kept pace. Further, operational analysis of key issues is still in flux. The Iridium/Cosmos collision event in 2009 demonstrated that there is a lack of data and tools available to all space actors needed to avoid major accidents. Additionally, the event demonstrated that the publicly available Two Line Element (TLE) sets are not sufficiently accurate to detect and prevent such incidents. Open source software is a relatively new trend in software development that rests on the principles of open collaboration. Unlike closed source software, the source code behind the software is publicly released and anyone is free to modify it. These modifications can become a completely new project or be integrated into the mainline development. Critically for the space surveillance application, software developed using open source software methods provides greater transparency, knowledge of what‟s inside the “black box”, flexibility, can reduce development costs, and can be used globally with fewer restrictions in regard to export controls and intellectual property restrictions. The focus of this paper is on the software aspects of moving the current legacy space situational awareness capabilities forward via an open source paradigm, so that all spacecraft operators have access to the basic tools needed to operate safely and efficiently in space. 1. INTRODUCTION The fundamental requirement of space situational awareness (SSA) is to provide actionable knowledge about events and activities in Earth orbit. A key component of SSA is space surveillance -- determining the present position of space objects and the ability to predict their future orbital paths. Related requirements are the detection of new space objects, the detection of spacecraft maneuvres, and the prediction of when one space object may interfere with another space object. Such interference may be physical in nature such as the February 2009 collision between the Iridium 33 and Cosmos 2251 or electromagnetic in nature such as the Galaxy 15 anomaly 1 . All of these requirements require space object ephemeris data; an Ephemeris is a table of predicted position and velocity at a sequence of times, usually at equal time intervals. The ephemeris data is generated by fitting mathematical models to tracking data. The tracking data includes data from radar and optical sensors. The radar sensors include phased array radars, dish radars, and fences. The orbit fitting process makes use of the residual between the actual measurement at time t and the computed measurement 2 at time t t   where the quantity t  is the timing bias. 1 For a more detailed description of the Galaxy 15 situation, see “Dealing With Galaxy 15: Zombiesats and On-orbit Servicing”, The Space Review, 24 May 2010, online at http://thespacereview.com/article1634/1 2 The computed measurements require an a priori estimate of the state vector.