Regression of GPS positions in C++ OpenCV or Eigen

Question

I am looking for an example of predicting/regression of GPS position at a given time based on a sparse data set.

Given N rows of [lat,lng,time ticks], I want to be able to ask what the lat, lng is at any given ticks that might not be present in the N rows.

Additional, is it possible to also get the derivatives at the same time such the orientation of the moving object can be found?

In OpenCV or Eigen that is.

What exactly is your coding problem? – user3791372 Mar 30 '16 at 00:21 — user3791372, Mar 30 '16 at 00:21

score 2 · Accepted Answer · edited May 23 '17 at 12:24

2

That is a linear regression problem.

Try using OpenCV's fitLine(). Go through the documentation and you should get a clear idea.

Try looking into this question

Once you get the output line function, a simple atan((y2-y1)/(x2-x1)) should give you the derivative.

edited May 23 '17 at 12:24

Community

answered Sep 05 '14 at 08:08

metsburg

1 Answers1