Slashdot Mirror
NASA Phoenix Mission Ready For Mars Landing
Several readers relayed the press release from JPL about the upcoming landing of NASA's Phoenix Mars Lander on May 25. It's going to set down in the north polar regions and look for indications of whether conditions have ever been favorable for microbial life. "Phoenix will enter the top of the Martian atmosphere at almost 21,000 kilometers per hour... In seven minutes, the spacecraft must complete a challenging sequence of events to slow to about 8 kilometers per hour... before its three legs reach the ground. Confirmation of the landing could come as early as 7:53 p.m. EDT. 'This is not a trip to grandma's house. Putting a spacecraft safely on Mars is hard and risky,' said Ed Weiler, associate administrator for NASA's Science Mission Directorate. 'Internationally, fewer than half the attempts have succeeded.'"
I am sure a lot of slashdot readers are interested (I know I was) in how does this beast actually look like. So here's a very good article on the Phoenix lander with a couple of fantastic artistic concepts based on the actual Phoenix.
"The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
NASA: Oh my, Mars, what big craters you have!
GrandMars: All the better to SWALLOW you with.. grrrr!
Do it yourself, because no one else will do it yourself. [beta blockade 10-17 Feb]
Because even at 8km/h you can do serious damage. Any lander has to be extremely light for takeoff from Earth and the transit to Mars, contain extremely fragile equipment, and end up there in one piece. "Bouncing" off Mars is not an option. That requires heavy, expensive materials, or some sort of complicated landed shield arrangement (e.g. giant inflatable bubble) that all add years of work and millions to the cost of the project. You could literally double or treble the cost of the entire project by "beefing up" the lander.
Plus, it has to land under autonomous control, so you really have no idea how fast it actually landed or exactly where until several minutes after it has landed - so coming in a little too fast isn't a good option, neither is a stray patch of rock (there are few "soft spots" on Mars, by the way - it's mostly rock). Much better to land as gently as you can manage and do your braking manoevures in the "air" as you come down. You've got plenty of time, the physics are easier to calculate, and there's less to go wrong.
The first few hours of a new lander's life on another planet are basically checking that everything still works, even with all the gentle landings in the world, things get broken that cost MILLIONS to put them up there. 50% of the things still never make it to the planet operational, even with all the good will in the world behind it. You want to spend MULTIPLES of the cost of the entire project on making the landings more difficult, more violent and less reliable when we can't even get half of what we send onto the planet successfully?
This is not a trip to grandma's house
You've never met my grandma. As a kid, going there felt like a 25,000 mph trip, and there are still skidmarks from my shoes trying wildy to decelerate while my parents dragged me into the house. And about half of the times they tried taking me there, it failed too...
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
#space on irc.freenode.net is hosting a Phoenix Landing Party on Sunday, May 25 to share in this momentous occasion for planetary exploration. We'll be following NASA TV through landing, then ogling the raw images when they are released several hours later. Historically, #space has been a hub for collaborative efforts in image processing by the space enthusiast community (Mars Exploration Rovers, Huygens, etc). Hope you can join us!
They're just learning from past mistakes.
Much like the experienced worker that estimates a month for a two hour job.
Yes. The details are hidden away on wikipedia where you'll never find them! Some details:
Mars 2 (1971): Landed but lost contact within minutes
Mars 3 (1971): Same
Viking 1 (1974): Landed and remained operational for 6 years
Viking 2 (1974): Landed and remained operational for 3 years
Phobos 1 (1988): Lost on the way to Mars
Phobos 2 (1988): Got into orbit, took some photos, then failed
The more recent ones you probably know about. To be fair, the Phobos 1 and 2 missions were planning to land on Phobos, not Mars, so maybe they don't count.
Wow, that isn't a fast transfer rate. That's about 1KB/s, 4KB/s, and 16KB/s, respectively. I guess you don't need too much more -- but still, I bet it's slower than they would like. The high resolution camera alone probably produces images that are a few megabytes in size. Let's say the images are like 4MB -- Transferring 4 MB at 1KB/s takes about an hour!
Given the slow xfer speeds and limited hardware they probably use -- I think it would be fun to be a programmer for NASA. That's one of the few applications where efficiency of communications, small memory footprint and efficient CPU usage probably still count for something.. I bet you everything they do when it comes to the software running on the lander tries to be as efficient as possible (especially communications-wise).
Also, isn't there something like an few minutes of latency for light to reach us from Mars? You can't even really do any really realtime interaction with the onboard computer on the Phoenix lander.. Imagine typing into a shell and waiting a minute for your characters to appear! Ouch! So I bet you they have to premeditate a lot of the changes they make to the software or operating environment way a head of time -- they probably just upload scripts of commands when updating the software or filesystem, etc.
I wonder how much freedom they give the people communicating with the lander. Do they triple-check every command sent to it to make sure noone does the inadvertent 'rm -fr
Cut them some slack! Most of us slashdot readers have trouble getting an Apache install right the first time through. These guys are doing nearly the impossible and they don't get much of a chance to fix any mistakes.
There are like THOUSANDS of possible things that could go wrong with the landing that DON'T because the engineers did their job. If you have ever engineered anything, you know how much you have to think ahead. They sat really hard and long and tried to perfect the landing process.
But it's darned hard. Mars is really really really far away. The data transfer speed to the lander is like 16KB/s on a good day. You can't send realtime flight data and have a pilot fly the thing with a joystick (because of the latency and the bandwidth is just too limited). You just have to build smart control logic into the thing and hope for the best.
And -- what can ruin the whole thins is -- just one largish rock in the wrong place and the whole mission is a failure. Historically, only 5 out of 13 landers made it to the surface operational!
So, stop being a douche and start appreciating how hard this all is. And it isn't just NASA -- the Brits also tried and failed. It's hard. NASA is doing a great job. Let's see you send 100LBS of spacecraft millions of miles away and have it get there safely. It's pretty amazing it ever worked at all!
Oh and what "corporate committes"? Last I checked NASA was a government agency.
Stop thinking like a corporate douch and start thinking like a scientist. These guys are smarter than you or I and give them some respect.
Here is a great scorecard of all missions to mars showing which have succeeded, which have failed, and why: http://www.bio.aps.anl.gov/~dgore/fun/PSL/marsscorecard.html
Space and Computers.