Alternate Turn Behavior at Waypoints

[GoogleCodeExporter]

APM currently waits until it crosses over the waypoint before it starts a turn 
to the next segment's heading.  This results in a guaranteed overshoot of the 
next segment's course line and subsequent hunting behavior to get back on 
course.

APM would perform better if the turn was started before the waypoint was 
actually passed.  The lead-in distance could be taken from the waypoint 
diameter, or (with more complexity) it could be calculated using the current 
groundspeed and turn rate.

This "leading the turn" is used in most GPS navigation systems from general 
aviation on up.

Thanks.

Original issue reported on code.google.com by ademar...@gmail.com on 18 Jan 2012 at 2:13

[GoogleCodeExporter]

Ping.

Original comment by ademar...@gmail.com on 12 Apr 2012 at 9:06

[GoogleCodeExporter]

Concur!  I've started working this in a clone, brnjones-exp, newWP branch.

Original comment by brnjo...@gmail.com on 12 Apr 2012 at 9:14

• Changed state: Started

[GoogleCodeExporter]

Original comment by brnjo...@gmail.com on 12 Apr 2012 at 9:17

• Added labels: Difficulty-Medium, Performance, Type-Enhancement

[GoogleCodeExporter]

Original comment by brnjo...@gmail.com on 12 Apr 2012 at 9:20

• Added labels: Navigation

[GoogleCodeExporter]

Nice!

Original comment by colinbou...@gmail.com on 12 Apr 2012 at 9:45

[GoogleCodeExporter]

Brandon-  Feel free to involve me any way you want in testing.  I have an 
APM1-XPlane setup and an APM2-EasyStar that flies regularly. 

Original comment by ademar...@gmail.com on 18 Apr 2012 at 4:51

[GoogleCodeExporter]

Yes, this would greatly improve a lot of things, including PID tuning, as less 
corrections would need to be made.

Original comment by mike.ben...@gmail.com on 19 Jun 2012 at 12:36