Fly-away Follow-Me fix

[kellyschrock]

Filter out spurious location updates using average speed
(vs. arbitrary speed limit).

Fixes #1128

[kellyschrock]

This solves the same issue as #1130, but uses average speed instead of an arbitrary 15 m/s (or other) speed limit.

[m4gr3d]

@kellyschrock is the constant factor used to calculate the upper limit for the current speed an arbitrary value?

[kellyschrock]

I originally used a much lower factor (1.25), and found that it took a very long time for the UAV to start following. In my experience, location jumping is only likely to happen very soon after starting Follow-Me mode (while the device isn't moving very fast if at all). In that case, anything less than a 4x increase in speed seems reasonable.

[m4gr3d]

So from your testing, is it behaving as expected with the changes in this PR.

[kellyschrock]

Yes, it is. I did several tests this morning, and it's behaving as expected.

On Tue, Oct 7, 2014 at 12:42 PM, Fredia Huya-Kouadio <
[email protected]> wrote:

So from your testing, is it behaving as expected with the changes in this
PR.

—
Reply to this email directly or view it on GitHub
#1145 (comment)
.

[m4gr3d]

Sounds good! The factor would probably need more testing with additional transportation means, but it's a great first step!
Could you store the factor into a constant variable, then I'll merge the PR.

[kellyschrock]

Will do. Sorry about all the commits, I had difficulty pushing to my fork.

On Tue, Oct 7, 2014 at 12:58 PM, Fredia Huya-Kouadio <
[email protected]> wrote:

Sounds good! The factor would probably need more testing with additional
transportation means, but it's a great first step!
Could you store the factor into a constant variable, then I'll merge the
PR.

—
Reply to this email directly or view it on GitHub
#1145 (comment)
.

[arthurbenemann]

You can always debase your branch and summarize the commits, and later do a
forced push to your fork.
On Oct 7, 2014 11:12 AM, "Kelly Schrock" [email protected] wrote:

Will do. Sorry about all the commits, I had difficulty pushing to my fork.

On Tue, Oct 7, 2014 at 12:58 PM, Fredia Huya-Kouadio <
[email protected]> wrote:

Sounds good! The factor would probably need more testing with additional
transportation means, but it's a great first step!
Could you store the factor into a constant variable, then I'll merge the
PR.

—
Reply to this email directly or view it on GitHub
<
https://github.com/DroidPlanner/droidplanner/pull/1145#issuecomment-58229327>

.

—
Reply to this email directly or view it on GitHub
#1145 (comment)
.