Could you please quantify "better than good enough"? As far as I know, bayesian models need to work with the logarithm of probabilities to get meaningful results for anything other than toy examples. Apart from that, are there any peer-reviewed papers that show a performance win with this concept for a real-world example?
You are missing that this is a 2-dimensional problem. Both wind velocity and boat velocity are 2d-vectors.
What is required is that the projection of the boat velocity in the direction of the wind is smaller than the wind. The magnitude of the boat velocity can always be greater as that of the wind as long as it does not travel in the same direction as the wind blows.
It's really simple: the clocks at the two checkpoints must be synchronized. The system probably uses ntp, and a GPS receiver as reference clock.
Here in austria, a system like this has been in use on motorways since 2003.
I never claimed sqlite is the only solution without adminstration, but I think this is the biggest advantage it brings to the table.
Also it seems to be very robust, and there are no license problems whatsoever. Its workings are very well documented. It is a very useful tool if you use it for the right tasks.