Impact of Launch Cadence on the Automation & Economics of Constellation Operations

Kurzfassung

When designing the operations concept for a satellite constellation a mission owner needs to consider the initial size of the operations team, when and how to ramp up the staffing, and ultimately the steady-state size of that team. In parallel to the routine operations of the existing constellation, the operations preparation and launch and early orbit phase for the next batch of satellites joining the constellation have to be performed, as well as the deorbit of degenerating satellites and the preparation of the constellation’s next generation. These factors have to be addressed at the design stage and trying to scale the approach from classical satellite operations to a constellation leads to the conclusion that several hundreds of operators would be required. This is cost ineffective, negatively impacting the ability to close a business case, nor credible because such a number of operators does not exist and cannot be trained up responsibly within the required time. The answer must consequently involve software and automation so that the workload of the operators is reduced. Several constellations are now operational where, for example, Starlink and OneWeb are already starting to refresh with a second generation. This paper compares the evolution of these constellations and reveals what can benefit those wishing to launch and operate their own constellation in future. We study how the launch cadence impacts the size and growth rate of the constellation and in turn the operations team. We uncover some unexpected trends that are common between operators, such as Planet, Spire, OneWeb and Starlink, and in turn the time and data operators have available for achieving the desired automation. We also discuss when machine learning or digital twins become feasible to employ and the type of mission phase they can help to automate during the lifecycle of the constellation. This also serves as a benchmark, where an assessment as to how much a new technology facilitating automation has to achieve in order for it to be felt in the total lifetime of the constellation. We conclude the paper with lessons that can be learned for upcoming constellations such as LEO-PNT.