Redundant Flight Computer and File Partition Maintenance Design for the Mars2020 Mission

The Mars2020 Perseverance rover is equipped with a variety of redundant avionics to ensure a reliable backup option in anomalous scenarios. Of major importance is the Rover Compute Element (RCE) which includes the primary RAD750 flight computer, nonvolatile memory storage, and in- terfaces to the other avionics assemblies such as the rover’s telecommunications and instruments. The rover is designed to autonomously switch to a backup RCE in a scenario where the prime RCE is continuously faulting, which occurred during the Mars Science Laboratory mission preceding Mars2020. Since the backup RCE could become the prime RCE at any time, it is important to regularly maintain critical files, parameters, and other aspects between the two RCEs that are not automatically synced. Additionally, each RCE contains a primary and backup nonvolatile memory partition. These are also unable to be autonomously synced but can be vital for contingency operations as also seen during the same MSL anomaly. This paper discusses the approach that was developed on the Mars2020 mission for syncing the RCEs at a regular cadence to ensure a safe and expedited return to nominal operations in a swapped RCE scenario. It will cover the original design intent of the redundant flight computer and backup file partitions and elaborate on the MSL anomaly where this design came to be of great use. Next, it will discuss various principles of developing a regular maintenance activity to sync each of these partitions in a way that is non-burdensome for the operations team to accommodate but places each backup partition into a safe and deterministic state. It will also include a discussion of some of the challenges involved in this process and finally illuminate some of the design recommendations for improving this redundancy design for future Martian surface missions like the Mars Sample Return (MSR) Sample Retrieval Lander (SRL).