Get bearing from weighted avergage

[letorbi]

Use sensors if everything else fails

[letorbi]

Well, it works now (200d9ce1f4ea934f81c699c7cf5b8e7a7a1738b4), but could still be improved. The bearing based calculation is quite clumsy. It might be a better idea to calculate a hypothetical last position from the average of the last measured positions.

An approach could look like this:

• calculate the vectors between all known positions
• normalize these vectors and create the average
• generate the hypothetical last position by adding the average to the current position

• calculate the bearing from the current position to the hypothetical position

  I leave the issue open until new improved calculation approach has been tested (and maybe implemented).

[letorbi]

The bearing calculation was improved by f07ab3cf63137090b9b65c22ecf8916774d015d6. However, the suggestion from the previous comment should be tested...