The third week in July there is going to be a pretest to make sure that everything is set up properly for Alliance members

NASA Mars Visualization Alliance

Meeting 4

Summary

June 4, 2002

8:30 a.m. PDT

In August there will be a FIDO (Field Integrated Design & Operations rover) test conducted by the MER (Mars Exploration Rover) engineering team to perform mission-relevant field trials that simulate mission operations scenarios. This test will last 10 days and incorporate a total of 20 Martian sols (two sols per day). Mars Viz members will be able to participate in the last day of the tests (August 19).

Some Alliance Members do plan to use the August FIDO test with members of the public so we would like to set up a pretest in July as a systems check for the FIDO tests in August, as folks at JPL likely will be a little too busy to troubleshoot for Mars Viz members during the actual FIDO tests. The pretest will use last year’s data from the FIDO field test. The pretest will last one day and incorporate 2 Martian sols. Before the pretest, everyone should make sure that they have access to the image products support page at http://marsdata.jpl.nasa.gov/muse/. To do this, they need to send their IP address to Anita Sohus of JPL.

Before the pretest, we would like to set up a File Exchange Interface (FEI) for the Mars Viz members. It is a subscription service that will allow members to have access to data. FEI will write the data directly into a directory that you specify, as soon as the data are available. The data is pushed to you as fast as possible, and pushed exactly where you’d like it to be. you actually can specify if you are only interested in particular parts of the data. You can customize it so that if you were only interested in the panorama camera or one particular camera, you could get only that if that is what you wanted. You also can have all of the imaging data.

The service can stay open for an indefinite amount of time. The FEI works only on a UNIX machine or Red Hat Linux on a PC. The best machine option is Sun Solaris followed by some other UNIX variety. It is not a Windows service but it is possible to get Linux Red Hat for Windows machines, which is free.

The client software is installed by a system administrator (SA) on the side of the museum. (Please sent name and contact info of this point of contact to Anita Sohus.) JPL will work directly with the museum SAs, one-on-one, to get the software running. It should take only a few hours for the SA to set up but is somewhat complicated. Each museum needs to identify a disk and establish their own directory structure. It is recommended that the disk be at least 100 gigabytes for the FIDO test in August. It is also highly recommended that members use JPL scripts to establish the directory tree so that information will be uniformly located regardless of where the tree head is located. As the data comes in from testing and for the actual mission it will be stored to the disk and available to the museum at any time. The client software is a search and service type and is aware each time new data come in, automatically shipping it to the subscriber.

The subscription service allows the user to specify only the portions of the data that are of interest. The user can customize the service so that it provides whatever the user wants. For example, if the user wishes only to have image data from a particular set of cameras the client can be installed to do that. The client can also be installed using time-based data, collecting only information on a given set of days.

Just to be very clear, this is independent of the Image Product Support page (http://marsdata.jpl.nasa.gov/fido/mer:fido/) for viewing. That doesn’t change. You can view things, no problem, both with Windows machines and with any machine that has Netscape and eventually Internet Explorer. But to have the subscription service be automatic, which is the only way you really want to get the data you need a UNIX machine. Then it is automatic, and you don’t have problems like buffer overflow and all kinds of stuff. It takes care of all the little niceties and writes the data to a disk for you.

We put a client there and you collect the data on your own disk locally. Then your SA can decide if you want to do some local pointing and things like that. We want to test two things. One, that you use the image product support page to look at the data as it is flowing. Two, that this subscription puts the data in a place where if you want to do more with it, you’ll have all the data. Then we’d like you to suggest any modifications in terms of how you can see it better or use it better. And, if as you get the real data and have it in a place, if you decide there is another product that the museums, can make with it, we’d all love to see that. That way everybody starts from the same place with respect to the data. It is all the same format and the same kind of directory structure. And it makes it simpler to compare notes when you want to talk about your product or your tool versus some other tool or product.

For testing and the actual mission, JPL has set up a separate, but mirror server to that of the Science Team so that bandwidth will not be an issue for the Alliance or for the Science Team. The data will be available as fast as it does for the Science Team.

JPL is running correlations on each of the image pairs and so there will be disparity maps that indicate the difference between a left and right eye. These should come out fairly soon. (For every stereo pair of images, the difference between each point that is in both images is identified. This is called disparity and gives an idea of the separation between left eye and right eye. Camera calibration is essential to this process).

Regarding models, the team has not standardized a format for that yet. Models could be exported in a VRML format to Alliance members. It is easy to run on Windows platforms.

All products are derived from the original data that is sent down to the users. Derived products include mosaics, color images and other products, but they all originate from black and white, single filter, one-eye images. The cameras are currently being calibrated.

During FIDO tests the Mars Outreach team will be working with particular scientists on writing image captions.

The American Museum of Natural History might use the FIDO test data with some of their summer class-type programs, with students, not the general public. The Exploratorium is investigating presenting the material to the public. Maryland Science Center would likely make it available to the public in one of their galleries in SpaceLink if the timing is right. But again, it is probably more worthwhile as a test to see how things run.