The Mars Reconnaissance Orbiter has seen the impact site. Burn was ten times too short, fell from 2-4 kilometers, almost-full tanks probably exploded on “landing.”
But they think it was a software error, which is good news.
10 thoughts on “Schiaparelli, RIP”
Comments are closed.
It’s rather impressive that we were able to locate the crash site of a lander, what, one day after it impacted?
Big scorch mark. Hard to miss.
Was that its Wile E. Coyote moment?
When are we going to stop bombarding Mars with our toxic wastes?
Seriously, does nobody do proper software testing for aerospace projects? This should have been simulated to death. Billions of dollars and years of effort wasted because these organizations are unable to learn from past mistakes.
SpaceX wont have this problem. They might have some other problem but their software is getting a lot of practice.
Generally I agree, sw QA is almost always underfunded, and most significantly, under time-budgeted in the project schedule.
We don’t know the details of what happened here … this failure might be such that all testing on Earth, or in sims would come back as “good”.
Current spacecraft development techniques do not have the benefit of flying often. This would be a very, very effective way to test. I imagine this particular failure mode would likely have been caught if a test model had been drop tested repeatedly. Doing so in Earth gravity and Earth atmosphere poses problems, but doing so on the Moon would be a fairly good analog (minus the parachute).
The giant paw print at the right side of the linked image gives one pause, I’ve seen my cats bat butterfly’s out of the air…
Jokes aside, maybe it’s time to give the probes a bit of intelligence, a neural net say, then train them in what’s wanted (a soft landing) rather than a single point of failure in a predetermined state engine or bit of ladder logic?
Granted I’m no rocket scientist or engineer, but it seems like it would be safer but more complicated to land from orbit instead of slamming full speed into the atmosphere and trying to assume how much braking you would need and get. I’m probably hopelessly naive.
In a Mars approach from Earth you’d have to slow down to get in orbit. That means either aerobraking or bringing more propellant – a lot more.
One interesting approach that’s been suggested is to enhance aerobraking by collecting some of the atmosphere the vehicle is traversing. This will require some care as the heat still has to be dissipated. The mass can then be used as reaction mass, either in electric engines or after conversion to chemical propellant.