ignore acceleration from driving ?

[skittlemittle]

since measurements are taken from a vehicle you'd have the acceleration spikes of driving which is not a measure of how bumpy the drive was (unless madmax).

gotta do something about that, do I

• only count quick short acceleration spikes, so we only count bumps and not speeding up?

• not listen to readings from the axis the forward acceleration is happening on(use compass and GPS to see which way the vehicle is moving in then remove that axis, or tell users to hold phone pointing forward?)

• remove acceleration from driving by checking velocity from GPS readings then removing that from the axis pointing forward before saving?

this might not even be a problem, who knows?

[skittlemittle]

you idiot, cars cant do more than one g and thatd be a smooth spike, a bump is a jagged chunk of spikes.

https://www.quora.com/What-g-forces-can-most-cars-produce