Expected navigation accuracy...

If I have an aircraft like the Bixler or Skywalker using APM 2.x, what is the expected navigation accuracy?

 

For example, if I want the plane to fly directly over a waypoint at a certain altitude, how much horizontal error can I expect 99% of the time? (i.e. would the closest approach be within +- 10m? +- 20m?, etc... 99% of the time).  What about altitude? 

 

Assume a long, stable lead-in course (plenty of time for APM to get things stable).  Assume winds are light and GPS fix is rock-solid.  I realize there probably is no hard and fast number out there, but what are people seeing?

 

I want to fly from a parking lot that has light poles, and I want to know how far apart the light poles really need to be to perform an auto-land reliably without hitting them.

 

 --Dan

 

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Developer

    I think the real answer is it depends on gps performance in your geographic location.  I would say for me (here) it would be +/- 5 meters at 95% confidence.  However, from past conversations with other users in different locations it appears that gps accuracy is not so good everywhere.  You need to get a feeling for your gps accuracy, as navigation accuracy depends on it.

  • Dan, you can autoland in a relatively narrow space, its the length you need for autolanding.

    If you setup the waypoints to bring it in on a straight line you should be able to land somewhe along a 10m wide corridor, i'd probably make it 20m wide if you can just to be safe and to account for wind.

    Also, don't commence the autoland waypoint until you are quite low (maybe 5m above the ground).

  • Developer

    I think you could expect a few meters in any direction as long as you give the plane enough time to line up. I would also add in some WPs to setup your approach.

    Jason

This reply was deleted.

Activity