Working on a tracking application using GPS It's all right, but sometimes due to closed areas or bad weather I get a wrong number when you plot them, so it does not look right with so many hops / jumps.
What algorithm should I run to filter bad signal, it looks like a blur algorithm app for me, but what do you think?
There are some options:
- Throw out external outsourced
- Use better GPS
- Use an external data source (photo on the road)
- I like to use a combination filter I - A Kleman Filter is a general (and often best) solution - it uses the amount of estimated average that is better than cheap IIR (Infinite Impulse Response). ) Filter:
Filtered value = Fill value * 0.75 + New value * 0.25
You can get GPS module which gives you 4-5 fixes per second, which You get the 'Cheap Feedback Time' filter above above.
You can get just a better GPS (SIRF III or better) which is not in the form of noise and not even the condition is the terraced indoor reception (where possible).
Consumers GPS units where "photo on the road" is possible, so errors from the road are not seen by the consumer, as well as some other techniques are also seen.
Implementing a hypothetical is not easy, but without the external dataset or sensor (such as the speed of the road), this is the best choice. Look at the code and tutorial for it.
- Adam
Comments
Post a Comment