COVARIANCE-BASED NETWORK TASKING OF OPTICAL SENSORS

Maintaining the catalog of Resident Space Objects (RSOs) is of critical importance to the protection of space assets. However, currently the Space Surveillance Network tasking for deep space RSOs is based upon an ad hoc RSO importance category system and only crudely accounts for the error in the catalog orbit estimates. TASMAN (Tasking Autonomous Sensors in a Multiple Application Network) is a comprehensive high-fidelity simulation environment of networked optical sensors that is designed to provide a flexible test-bed for dynamic and responsive mission planning algorithms. Simulations performed using TASMAN show the results of exploiting the RSO state error covariance, a quantity already computed centrally, to more effectively schedule the sensors to reduce error in the catalog states. Significant improvements in the median catalog accuracy are apparent from using covariance-based scheduling.