Accident and incident investigation for space-based systems
Space is a ‘high-risk’ business; yet many missions fail to learn from previous failures.
This can be illustrated by the common causes between the Mars Polar Lander and Beagle 2 missions. It is very difficult to identify these common causes and to anticipate that they might affect a future mission. Space systems often fail from complex interactions between software issues, hardware faults, navigation and environmental constraints etc.
Together with teams from NASA, the European Space Agency and the US Air Force we have worked for 20 years to identify techniques that assist in mishap investigations for space based systems. The intention is to identify a wide range of root causes and contributory factors so that engineers can learn the lessons of previous failures.
Much of this research was motivated by the Columbia and Challenger tragedies.
A key requirement has been to develop analytical tools that can trace how faults develop between multiple private companies who increasingly must work together on complex space missions, following the Obama funding reforms.
Why is this research important?
We have worked with the NASA Shuttle and International Space Station teams and the ESA human space flight groups. An innovative feature of our work is that we are using information about previous failures to assess risks for future missions. For Mars missions, we built software to help the crew cope with on-board systems faults during the 90 minutes that it would take radio communication to reach ground control. These techniques can also be used to identify failure modes for conventional satellite operations.