Integrated tasking, processing, and orbit determination for optical sensors in a space situational awareness framework

Space situational awareness;Optical sensors;Tasking;Orbit estimation;Computer vision

Murphy, Timothy S. ; Holzinger, Marcus J. Aerospace Engineering Johnson, Eric Lightsey, Glen Flewelling, Brien Sabol, Chris ; Holzinger, Marcus J.

University:Georgia Institute of Technology

Department:Aerospace Engineering

关键词: Space situational awareness;    Optical sensors;    Tasking;    Orbit estimation;    Computer vision;   

Others  :  https://smartech.gatech.edu/bitstream/1853/59930/1/MURPHY-DISSERTATION-2018.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

This dissertation explores the following. Coordinating a large network of electo-optical sensors (EOS) for an effective Space Surveillance Network requires several novel capabilities. The first piece of this work involves the search set, or the set of orbits which may contain relevant object(s). A tasking algorithm is needed to optimally choose a trajectory through the sky that a sensor should take to search a set of orbits. Because of the malleable definition of a set, this allows EOS to be tasked on a wide variety of search and reacquisition problems. Next, when taking actual data the sensor exposure time, slew rate, and campaign length need to be chosen to optimize the quality of image data. These tasking parameters are chosen with respect to the detection and estimation algorithms themselves, which all relate back to a maximum likelihood method. Next, the object detection algorithms should be as sensitive as possible. This enables a larger network of lower cost telescopes to be deployed, and ensures that performance is robust to light pollution, enabling new telescope locations. These types of networks are needed to allow the kinds of sensor architectures which support interesting handoff and reacquisition problems. Finally, to make proper telescope communication, hand-off, and long term reacquisition possible, detection algorithms should utilize any prior information (search set) on a particular object or class of objects for more sensitive and efficient detection. This supports hand-off between arbitrary locations and longer delay times before reacquisition from the detection side of the problem.