Re: New Flight Prediction Program [1 Attachment]

Jerry
 

I have made a major update to the flight predictor program.  You can still find it here: http://www.s3research.com/tools/flightpredictorHere are the major changes. 
 
  • The weather data page is gone.  The program now downloads the wind data as it runs the prediction.  This means it will take some time to run long predictions.  Additional predictions done won't reload the wind data.
  • I have added more wind points.  It now uses 46 points from 1mb (150K feet) to 1000mb.  The interpolation between points has also been improved.
  • The above 2 improvements were needed to support floaters.  This is now working fairly well now.  I generated this aprs.fi overlay for M0XER-6 a little over 24 hours ago. So far it is under 100miles off the predicted path.  I think this is more from a stable balloon and weather than the prediction.  Here is another I did for M0XER-3 this afternoon.
  Known issues:
  • I wouldn't try to do a prediction over 100 hours (6000 minutes).  There are some memory issues that pop up if you try much more than that.
  • It looks like I have a boundary issue at latitude +180/-180 boundary.  At this point that probably only affects Leo.
  • I have had a request to be able to use your own weather data.  I haven't implemented that yet.
Give it a try and let me know what you think.  If you see something that needs to change or be improved let me know.
 
Jerry Gable
Balloon Flights from APRS-IS
Launch Notifications by email, twitter, & APRS
http://www.s3research.com/flightdata/


From: "Jerry Gable jerrygable@... [GPSL]"
To: GPSL List <gpsl@...>
Sent: Wednesday, July 23, 2014 9:50 AM
Subject: [GPSL] New Flight Prediction Program [1 Attachment]

 
[Attachment(s) from Jerry Gable included below]
Over the past couple of months I have been talking with Keith Kaiser about doing a new flight prediction program.  We think it is to a point where we need to get some more people looking at it.  I still consider it Beta software but appears to be pretty functional.

Here are the key features:
  • Downloads a variable size portion of the GFS 0.5 x 0.5 forecast for the prediction.  This is 42 pressures from 10mb to 1000mb.
  • Create multiple predictions and display them on the same map (screen capture attached).
  • Load csv files of a flight onto the map as well.  Currently understands aprs.fi and findu.com export formats.
  • Displays altitude graph of the predictions and flights.
  • Movable "Ruler" that allows you to measure distances.
  • Upload a prediction overlay to aprs.fi (example the same as the attached screenshot: aprs.fi/#!kml=http...).
  • Support for short "Floater" flights.  Does not support longer  (>4 to 6 hours) because only one forecast time is included in the weatherdata.

There are still some things that need to be done but I wanted to get some opinions on the tool before I call it done.  The things I know that need to be done are:
  • There are very few checks on the data.  You can enter values that will cause problems.
  • It currently has issues on Android and IOS tablets.
You can find the tool here: http://s3research.com/tools/flightpredictor/.  There is a list of the overlays that have been uploaded here: http://s3research.com/tools/overlays/.  At this point the overlays are all my test cases.

If you try it let me know what you think, good or bad.
 
 
Jerry Gable
Balloon Flights from APRS-IS
Launch Notifications by email, twitter, & APRS
http://www.s3research.com/flightdata/


Join GPSL@groups.io to automatically receive all group messages.