Is software the cause of the Schiaparelli ExoMars lander crash? -Jordans Thoughts

About a month ago on the 19th of October 2016 the European & Russian space agencies joint ExoMars Schiaparelli mission took a turn for the worst when entering the martian atmosphere for its landing, after spending 7 months just trying to reach the red planet. The lander crashed after air braking and successfully deploying its parachutes + heat shields. It was upon prematurely ejecting the parachute and heat shields, then attempting to use its engines to slow it down for touchdown, that the mission went wrong. The 3 hydrazine engines cutout while the lander was still 3.7km above the surface causing the lander to free fall into the ground, creating a 2 meter wide crater on the surface of mars. The cause has still yet to be determined by ESA, however since knowing of the crash I did not suspect it was cause by a hardware error but a software error instead, which has been suggested by the latest report.

The crash site was photographed by NASA’s Mars Reconnaissance Orbiter (orientated so North is up).

My reasoning is based on history repeating itself, and concepts on hardware and software failures I had learnt from taking an Introduction to Aerospace Engineering course earlier this year. Looking at the past of Mars landers I noticed NASA’s Mars polar lander had a very similar event happen to it in 1999. The polar lander had successfully deployed its parachute and heat shield in the atmosphere, but upon using its engines to slow decent they stopped firing and the lander fell 40 meters before hitting the ground. The cause was not because of engine failure but because of the software. As the landing legs were deploying their movement shuddered the lander which exerted a force on the accelerometers (devices that measure acceleration on the landing legs). This ‘shuddering’ caused a false signal from the accelerometers saying a landing leg had hit the ground, despite being 40 meters in the air. The output of the accelerometers was made priority by the software over of the output of the other devices on the lander (like the altitude detectors) and the spacecraft cut the engines as it believed the legs had hit the ground. All the instrument parts worked as they were designed, but it was poor software and its interpretation of the signals that caused the engines to turn off and make the lander plummet into the ground.

Just 2 days ago (23/11/2016) ESA released an update on the investigation which hinted at an issue with software. It was revealed that all hardware components that were vital for landing were functioning correctly at the time the lander was descending under the parachute. However once the lander had ejected from the parachute the Inertial Measurement Unit (IMU), a device that measures rotational acceleration, hit its maximum value. This then sent an output signal to the navigation control computer, which the computers software interpreted as meaning the altitude of the lander was negative (the lander thought it was underground). Hence the software told the parachutes and heat shield to eject early and only fire the decent engines for a short period of time, leaving the lander still 3.7km high and free to fall. This is still under investigation and there aren’t too many details, but this is the most likely cause revealed thus far.

There are still other unanswered questions which hopefully will be answered by ESA by early 2017. The image above shows impact crater created by the spacecraft and the locations of where the ejected heat shields and parachute landed. The impact crater shows something odd which doesn’t make immediate sense if the lander fell directly down to the ground (as it should have). More impact debris (dark material) can be seen to the West (Left) of the crater than the East (Right). A pattern such as that on a impact crater suggests the lander hit the ground at a very low angle to the surface, as opposed to a perpendicular angle. This and other questions like ‘what caused the arc of dark material to the North East of the crater’ have still to be answered. Maybe the debris patterns are just due to some explosions of propellant that occurred when the lander hit the ground, or evidence of some rotational movement of the lander as it hit the ground.

Thankfully the mission was not of great scientific importance, with the mission being more of a test by the European and Russian space agency to show their capability to land large probes on the surface of Mars. However, this mission did hold several scientific instruments (mainly meteorological instruments) and was set to be the first to look at atmospheric electromagnetic fields/forces from the surface of Mars and what influence these may have on the dust within the atmosphere (potentially giving insight into the cause of martian dust storms). It is a setback for the ExoMars program, but only 16 of the total 44 missions to Mars in history have been completely successful so it is fair to say it’s not easy. Hopefully we understand these issues better soon, because if we plan on doing a manned mission to Mars by 2040 then we will need to get better at landing things.




-These are the things I think about in my spare time or when i’m trying to avoid deadlines for university.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s