Software designed to give spacecraft more autonomy could support a future where swarms of satellites navigate and complete scientific objectives with limited human intervention. This initiative, known as the Distributed Spacecraft Autonomy (DSA) project, is being spearheaded by NASA’s Ames Research Center in California’s Silicon Valley. Caleb Adams, the project manager, oversees testing alongside racks containing 100 spacecraft computers, aiming to enhance the adaptability of multi-spacecraft missions.
The DSA project is developing software to efficiently allocate tasks between spacecraft using ad-hoc networking and enable human-swarm commanding of distributed space missions. This advancement is crucial as astronauts living and working on the Moon and Mars will rely on satellites for navigation, weather, and communications relays. Automating satellite communications will allow explorers to focus on critical tasks instead of manually operating satellites.
Addressing Communication Delays
Long-duration space missions require seamless coordination between systems on Earth and other planets. Satellites orbiting the Moon, Mars, or other distant areas face communication delays with ground operators, which could limit mission efficiency. The DSA project offers a solution by testing how shared autonomy across distributed spacecraft missions can make spacecraft swarms more self-sufficient, allowing them to make decisions and adapt to changes with minimal human intervention.
By adding autonomy to satellites, they become capable of providing services without waiting for commands from ground operators. Distributing autonomy across multiple satellites enables them to operate like a swarm, effectively giving them a “shared brain” to accomplish goals they couldn’t achieve alone.
Sharing the Workload
The first in-space demonstration of DSA began onboard the Starling spacecraft swarm, a group of four small satellites demonstrating various swarm technologies since July 2023. The Starling mission serves as a testing and validation platform for autonomous swarm operations. The swarm first used DSA to optimize scientific observations, autonomously deciding what to observe without pre-programmed instructions. These autonomous observations led to measurements that could have been missed if an operator had to instruct each satellite individually.
The Starling swarm measured the electron content of plasma between each spacecraft and GPS satellites to capture rapidly changing phenomena in Earth’s ionosphere. The DSA software allowed the swarm to independently decide what to study and how to distribute the workload across the four spacecraft. If one member of the swarm was unable to complete its task, the others could react and fulfill the mission’s goals.
The Starling 1.0 demonstration achieved several firsts, including the first fully distributed autonomous operation of multiple spacecraft and the first use of space-to-space communications to autonomously share status information.
A Helping Hand in Orbit
Following the successful demonstration on Starling 1.0, the team explored additional opportunities to use the software to support satellite swarm health and efficiency. Continued testing of DSA on Starling’s extended mission included PLEXIL (Plan Execution Interchange Language), a NASA-developed programming language designed for reliable and flexible automation of complex spacecraft operations.
Onboard Starling, the PLEXIL application demonstrated autonomous maintenance, allowing the swarm to manage normal spacecraft operations, correct issues, or distribute software updates across individual spacecraft. Enhanced autonomy makes swarm operation in deep space feasible, reducing the need for constant communication with Earth, which can take minutes or hours depending on distance.
Simulated Lunar Swarming
To understand the scalability of DSA, the team used ground-based flight computers to simulate a lunar swarm of virtual small spacecraft. These simulations demonstrated swarms providing position, navigation, and timing services on the Moon, akin to GPS services on Earth. The DSA team conducted nearly one hundred tests over two years, demonstrating swarms of different sizes at high and low lunar orbits.
The lessons learned from these early tests laid the groundwork for additional scalability studies. The second round of testing, set to begin in 2026, will demonstrate even larger swarms, using flight computers that could later go into orbit with DSA software onboard.
The Future of Spacecraft Swarms
Orbital and simulated tests of DSA are paving the way for increased use of distributed autonomy across spacecraft swarms. Developing and proving these technologies increases efficiency, decreases costs, and enhances NASA’s capabilities, opening the door to autonomous spacecraft swarms supporting missions to the Moon, Mars, and beyond.
Milestones:
- October 2018: DSA project development begins.
- April 2020: Lunar position, navigation, and timing (LPNT) simulation demonstration development begins.
- July 2023: DSA launches onboard the Starling spacecraft swarm.
- March 2024: DSA experiments onboard Starling reach the necessary criteria for success.
- July 2024: DSA software development begins for the Starling 1.5+ mission extension.
- September 2024: LPNT simulation demonstration concludes successfully.
- October 2024: DSA’s extended mission as part of Starling 1.5+ begins.
NASA Ames leads the Distributed Spacecraft Autonomy and Starling projects, with funding from NASA’s Game Changing Development program within the Space Technology Mission Directorate. The Small Spacecraft Technology program within the same directorate funds and manages the Starling mission and the DSA project.
