Ingenuity Comes To The Rescue Of Mars Sample Return
"Two Ingenuity class helicopters will be used as back-ups to transfer samples to #Mars sample return ascent vehicle – @Dr_ThomasZ #Astrobiology
—
FYI this redundancy option would not even be possible if @sciastro John Grunsfeld had insisted that Ingenuity be sent to Mars. pic.twitter.com/UOvAcLmFrJ
— NASA Watch (@NASAWatch) July 27, 2022
NASA is at its best when they try bold experiments, like Ingenuity.
Yes, but the next two helicopters on the sample return mission won’t be bold experiments. They’ll probably be two or three times more expensive and less capable that Ingenuity. They will be critical parts of NASA’s highest priority planetary mission. That means all the things NASA expects from a class A mission. So no off the shelf commercial parts like a Snapdragon 801 processor, and lots and lots of additional testing and design reviews. Remember, the bold experiment with Ingenuity was only possible because it was considered a engineering test and failure was an option.
Think they will just use the obsolete Snapdragon 801 processor again. It is proven working hardware on Mars. Besides there isn’t a rad-harden processor chip that have the capability to control a helo drone and stream video.
I know people who have tried to do that for flight instruments on NASA missions. Specifically using a chip which had flown on previous non-NASA spacecraft but which was not technically certified for the flight environment in question. The argument was that it had been proven to work in that environment, even if it hadn’t been tested and formally put on the approved parts list for a flight mission. The answer was no. The flight experience was not relevant because they might just have gotten lucky. Actually, gotten lucky multiple times. That’s how NASA treats hardware for flagship missions.
So either they’ll try to certify the Snapdragon 801 (which may not be possible) at great expense, or they find some alternative (even if it means reducing capabilities, like streaming video, which isn’t mission critical) or someone will write a waiver saying it’s ok to risk the whole Mars Sample Return project on an uncertified part (which no project manager would want to do.)
Ability to stream video is useful since that means the drone is broadcasting data while it is flying. Of course that means you need something like the Snapdragon processor on the lander to processes the data.
Rad-harden version of the Power-PC chip is too antiquated and with not enough computing power for most current consumer applications for gadgets. Not adequate as drone flight control hardware.
That could be why they are talking about sending two of them. NASA rarely does that kind of redundancy for flagship missions. One break in the usual pattern could indicate another.
The radhard BAE 5545 might do…
Maybe not. It is still a 45nm die quad-core processor running at 466 MHz! Versus the obsolete Snapdragon 801 MSM8974AB 28nm die quad-core processor running at 2.36 GHz. Plus someone have to write the flight control software for the BAE 5545 processor.
Maybe having a few backup Snapdragon processors in the helo fetch drone will do for redundancy.
According to wikipedia development cost was $80M… so while a tech demo, certainly JPL attempted to retire a significant amount of risk. Also according to their 2018 paper “At the heart of the helicopter avionics is a Field-Programmable Gate Array (FPGA). The FPGA implements the custom digital functions not implemented in software due to resource limitations of the processors (e.g. I/O or bandwidth limits), timing requirements, power considerations, or fault tolerance considerations.” So to focus on the Snapdragon is a bit misleading. Undoubtedly as a mission critical system the price tag will go up substantially… I’d guess about 2x.
..