MRO Orbital Data

[Arrowstar]

Hi all,

I'm in the middle of putting together a proposal for a project I'd like to conduct for an advanced fluid mechanics course I'll be taking this upcoming quarter. What I'd like to do is put together some simulations of aerocapture/aerobraking heating at Mars, and I'd like to focus on the Mars Reconnaissance Orbiter to begin. In order to accomplish this, I'm going to need some data on MRO's trajectory. I've already been to JPL's HORIZONS website for their ephemerides data, but I'm finding that their data stream ends on March 12, the day after Mars orbital insertion. I was hoping for some data on the 6 months of aerobraking that MRO did after that, before the science phase started in November.

Does anyone know where I might find some additional information in this regard?

Thanks!:tiphat:

 

[tblaxland]

No trajectory data but I did find this document that should be of interest to you: MARS RECONNAISSANCE ORBITER OPERATIONAL AEROBRAKING PHASE ASSESSMENT

 

[tblaxland]

This should be what you want, SPICE files from launch to 1 Jan 2010: http://starbrite.jpl.nasa.gov/pds/viewProfile.jsp?dsid=MRO-M-SPICE-6-V1.0

 

[Arrowstar]

I'll have to look into those. I grabbed the SPICE toolkit for MATLAB, now I just have to see if I can't weasel some trajectory information out of all of this. Thanks!

Btw, the paper you presented was also excellent and complements another I found from JPL nicely.

---------- Post added 03-05-10 at 12:46 AM ---------- Previous post was 03-04-10 at 08:22 PM ----------

Alright, I think I got the SPICE-MATLAB interface working after a few hours. My current results, in the form of the MRO trajectory at the start of MRO aerobraking (plus a few orbits):

So far, so good! The next step is to find and implement a Mars atmospheric model so I can start looking at heating. This will clearly require some research before any coding takes place...

 

[tblaxland]

Looks good :thumbup: A bit of Googling shows up some papers on MRO aero-thermal modelling but I'm too far out of my depth to know what would be useful to you.

BTW, that graph would be very intriguing if you could colour code the plot by date, like here:

Don't let me distract you from more important work though

 

[Arrowstar]

Looks good :thumbup: A bit of Googling shows up some papers on MRO aero-thermal modelling but I'm too far out of my depth to know what would be useful to you.

No worries, I suspect I'm basically out of my league at the moment too. Of course, my goal is to put together a heating model that's not derived straight out of some paper published by the MRO team. I want to see if I can't get close to their results (which is what the papers are useful for) using what I know and what I expect to learn in this upcoming quarter. If nothing else, it'll be a good experience, but the Martian atmosphere is a complicated animal to tame, as well. We'll see.

Since the goal of this project isn't orbital mechanics so much as it is aero/thermodyanmics, I'm going to hold off on coloring the plot. But if I feel so inclined and bored one day, maybe I will.

---------- Post added at 08:24 PM ---------- Previous post was at 12:17 AM ----------

Okay, so while I'm waiting to figure out some of the aero/thermo for this little project, I decided I could go ahead and use some of that ephemeris data that JPL provided to create a nice little solar system simulator. (Some may recall that I attempted this a bit back, but I didn't have this JPL data to start off with.) Here's a shot of the inner solar system from the date of MRO's launch to just after Mars intercept. Keep in mind that only each body's initial state (position and velocity) is taken from the JPL data, from the date/time that I wish it to be taken from. The trajectories you see are all calculated.

The red line in a transfer orbit is MRO, and as I said, only it's departure velocity and position were taken from navigation data.

This is all well and good, but I do have one little problem with my numerics. Here's a picture of Mercury's orbit (the units on these graphs are kilometers, and yes, that is times ten to the seventh power):

Some of you may be able to spot the problem, but let's zoom in slightly:

The orbit trace lines are separating, clearly. Now, I say it's my numerics because I did a quick grid-independence study and I find the error between orbits shrinking as time step decreases, though admittedly I can't make my time steps too small or the simulation will take too long to complete. (For those interested, the graphs here were generated with a 100 second time step, which is a very small chunk of time to split 8 months into!)

At the moment, I am just using a simple Explicit Euler scheme to determine states in the next time step. I did try using a fourth order backwards difference formulation, but I didn't notice any appreciable change in my results.

So here are my two questions. First, keeping in mind that I'm basically numerically integrating Newton's second law, is there a better suited integration scheme I can use? Second, and related to the first, might there be any way I can decrease the required computation time without giving up (significant) accuracy? I suspect the answer here is no as CPU time is based on time step size, but I figure it's worth asking on the off chance someone can suggest a tip or two.

Thanks!