Design and Validation of an Autonomous Mission Manager towards Coordinated Multi-Spacecraft Missions
Session
Technical Session 12: Constellation Missions
Location
Utah State University, Logan, UT
Abstract
For ambitious upcoming aerospace missions, autonomy will play a crucial role in achieving complex mission goals and reducing the burden for ground operations. Standalone spacecraft can leverage autonomy concepts to optimize data collection and ensure robust operation. For spacecraft clusters, autonomy can additionally provide a feasible method of ensuring coordination through onboard peer-to-peer scheduling. However, in exchange for providing flexible mission capabilities and operational convenience, autonomy introduces additional uncertainty and software complexity, which complicates the mission assurance process. This research presents a framework for designing and testing schedules consisting of heavily constrained tasks.
The core of this framework, the Schedule Manager (SM), manages tasks by associating constraints with each task including time windows, task priority, conflict categories, and resource requirements, which assures that tasks will only run when capable. This increased control over individual tasks also
improves the modularity of the overall mission plan, and provides a built-in fail-safe in the event of unexpected task failure through the loading of predefined contingency schedules.
The SM can use estimated task durations and resource requirements to simulate schedules ahead of time, which can be used on the ground for schedule validation and onboard as a method of prognostics and to calculate resource availability windows. The ability to predict availability windows onboard and dynamically adjust depending upon currently scheduled tasks enables peer-to-peer tasking and scheduling. For example, a spacecraft can schedule a coordinated action by broadcasting the task requirements in an availability window request to all applicable spacecraft. Then, based upon the availability windows received from each spacecraft, the coordinating spacecraft can then issue a final task scheduling command with a much lower probability of conflict. The SM has been integrated with the core Flight System (cFS) from NASA, which has flight heritage on previous successful large-scale missions such as the Lunar and Dust Environment Explorer (LADEE). This integration is in the form of a cFS application called the cFS Schedule Manager (CSM), which will manage the operations for the Space Test Program Houston 7 Configurable and Autonomous Sensor Processing Research (STP-H7-CASPR) experiment that is planned for launch on SpaceX-24 to the International SpaceStation (ISS) in December 2021. Software validation was achieved with cFS unit tests, functional tests, and code analysis tools. Demonstrations were built using the COSMOS ground station and the 42 spacecraft simulator, and these were tested with a cluster of development boards in the loop as representative flight hardware.
Design and Validation of an Autonomous Mission Manager towards Coordinated Multi-Spacecraft Missions
Utah State University, Logan, UT
For ambitious upcoming aerospace missions, autonomy will play a crucial role in achieving complex mission goals and reducing the burden for ground operations. Standalone spacecraft can leverage autonomy concepts to optimize data collection and ensure robust operation. For spacecraft clusters, autonomy can additionally provide a feasible method of ensuring coordination through onboard peer-to-peer scheduling. However, in exchange for providing flexible mission capabilities and operational convenience, autonomy introduces additional uncertainty and software complexity, which complicates the mission assurance process. This research presents a framework for designing and testing schedules consisting of heavily constrained tasks.
The core of this framework, the Schedule Manager (SM), manages tasks by associating constraints with each task including time windows, task priority, conflict categories, and resource requirements, which assures that tasks will only run when capable. This increased control over individual tasks also
improves the modularity of the overall mission plan, and provides a built-in fail-safe in the event of unexpected task failure through the loading of predefined contingency schedules.
The SM can use estimated task durations and resource requirements to simulate schedules ahead of time, which can be used on the ground for schedule validation and onboard as a method of prognostics and to calculate resource availability windows. The ability to predict availability windows onboard and dynamically adjust depending upon currently scheduled tasks enables peer-to-peer tasking and scheduling. For example, a spacecraft can schedule a coordinated action by broadcasting the task requirements in an availability window request to all applicable spacecraft. Then, based upon the availability windows received from each spacecraft, the coordinating spacecraft can then issue a final task scheduling command with a much lower probability of conflict. The SM has been integrated with the core Flight System (cFS) from NASA, which has flight heritage on previous successful large-scale missions such as the Lunar and Dust Environment Explorer (LADEE). This integration is in the form of a cFS application called the cFS Schedule Manager (CSM), which will manage the operations for the Space Test Program Houston 7 Configurable and Autonomous Sensor Processing Research (STP-H7-CASPR) experiment that is planned for launch on SpaceX-24 to the International SpaceStation (ISS) in December 2021. Software validation was achieved with cFS unit tests, functional tests, and code analysis tools. Demonstrations were built using the COSMOS ground station and the 42 spacecraft simulator, and these were tested with a cluster of development boards in the loop as representative flight hardware.
