undefined | Better HN

0 pointsThatGeoGuy12y ago0 comments

On the topic of not being aware of the "spoofability" of GPS, it is interesting to note that many of the older Inertial Navigation Systems (GPS + IMU) use GPS to check against the accelerometers and gyroscopes in the IMU. Effectively, they would throw out the results from the internal dead-reckoning system if it didn't agree with GPS, instead of the other way around!

I guess it has more to do with the fact that unless you have a great Kalmann filter design, your IMU will likely drift off course rather rapidly, whereas GPS spoofing wasn't as easy or as popular as it is today (and I would say it really isn't that popular outside of major areas today, as it stands). Either way, you're right, the GPS coordinates should definitely be checked against the internal dead-reckoning. However, then you have to ask yourself how you know the internal dead-reckoning is still on course. It's a tricky problem, and hopefully the solution doesn't just tend towards "add more sensors."

EDIT: changed an "isn't" to "wasn't"

0 comments

1 comments · 1 top-level

mikeash12y ago

In such a system, what's the point of having the IMU at all? Is it just for the case when GPS becomes completely unavailable?

If I were designing such a system (and I'm sure I've overlooked about a million subtleties), I imagine I would make it so that the GPS corrects the IMU for small perturbations that lie roughly within the IMU's predicted error budget, and that the IMU causes the GPS to be ignored if the GPS coordinates suddenly diverge greatly.

In other words, the IMU is going to be saying something like, "our position is X, to within 5km". If the GPS disagrees with X by 1km or perhaps even 10km, correct the IMU's current state using the GPS. If the GPS disagrees with X by 100km, ignore the GPS until it gets its act together.

Seems like a decent first pass, at least.

j / k navigate · click thread line to collapse