A reliability estimation tool for reducing infant mortality in Cubesat missions

Abstract

For many years, traditional satellite design philosophy was dominated by highly reliable components, conservative designs and extensive performance testing at subsystem and integrated system levels to achieve long lifetimes in the harsh space environment. CubeSats attempted to choose a different philosophy, utilizing suitable state-of the art, commercial-off-the shelf products, yielding, if successful, an increased performance per mass figure of merit for those small vessels at potentially higher risk but lower cost. CubeSats seemed to promise universities and companies to be faster, better and cheaper - once more in history. Unfortunately, many CubeSat missions, especially university-built ones, never achieved a detectable functional state or failed shortly after the satellites were ejected from their deployer. Data based on our developed CubeSat Failure Database (CFD) and research carried out by others suggest, that a great percentage of those early failure cases could have been detected and avoided by more careful and adequate system-level functional testing on the ground. However, many university teams still fail to plan with adequate resources for system level functional testing or are confronted with hard deadlines, thus unable to complete appropriate integrated system testing on a sufficient level, and launching a satellite that never was adequately functional. Ongoing work on a novel reliability estimation tool using Bayesian methods is introduced to fill this gap and to provide meaningful data for all developers on the achievable reliability and required functional testing time of their CubeSats. Using test data and reliability goals for their actual mission, merging that data with statistical data from past missions and a database of subjective developer's beliefs, CubeSat developers should now be able to estimate their required functional testing time on subsystem and system level at an early project stage, as a function of the targeted reliability goal for their CubeSat. Alternatively, if the required resources (testing time, money, knowledge) are not available, CubeSat developers and program managers can still use the tool to now quantify a resulting realistic lower boundary for the expected system reliability of the mission, and decide, if their mission goals can be fulfilled or not with a certain probability. To evolve CubeSats into more reliable and accepted platforms for scientific payloads and commercial applications, it is utmost important to avoid or reduce the many infant mortality cases, where no or little useful data is produced by the satellite. To guide developers towards higher success rates without losing the spirit of using novel, state of the art technology in fast mission timelines, the reliability estimation tool should ensure higher reliability of CubeSat missions without drawing too much resources nor imposing too many burdens on the CubeSat teams.