No abstract

In this work, we explore the concept of a secondary "data mule" consisting of a small satellite used to ferry data from a Mars mission to Earth for downlink. The concept exploits the fact that two nearby optical communicators can achieve extremely high data rates, and that a class of trajectories called "cyclers" can carry a satellite between Mars and Earth regularly. By exploiting cycler orbits, the courier needs minimal onboard propulsion.However, cycler orbits have long periodicity, as it can take years for the satellite, Mars, and Earth to repeat their relative geometry. Therefore, we propose the use of a network of such cycler "couriers" on phase-shifted trajectories to achieve a regular cadence of downlink trips. We design a series of search and optimization steps that can output a set of trajectories that at first approximation have low onboard propulsion requirements and can be used for any regular logistics network to and from Mars, then derive the link budget for proximity optical communications to show that this network can ferry large amounts of data.

No abstract

We present a final overview of the efforts by the Naval Air Warfare Center Weapons Division (NAWCWD) and the Jet Propulsion Laboratory to automate the operation of the largest fleet of autonomous maritime vehicles. The vehicles are intended for large-scale demonstrations and tests of US Navy systems or tactics. This review covers a preexisting distributed architecture for human-in-the-loop control of several autonomous high-speed boats with a main focus on the planning, formation assignment, and formation switching pipeline. Algorithm capabilities are described and validated in simulation and field tests using real-world vehicles. Theoretical lower bounds on the time required to change formations are also derived and used to bound our experimental performance. We are able to present data from the 2019 “final exam” that pitted this architecture in a head-to-head competition, which was won after a perfect run with no hazardous maneuvers recorded under autonomous control. The effort concluded successfully on December 31, 2020, with the delivery of code and documentation after successful integration and testing exercises in 2019 and 2020. We conclude the paper with discussions of limitations, extensions, and suggestions for future work.