Hacker News Reader

Comment by eckza | Hacker News Reader

Probably not, as it was only optioned to the Accord and Vigor, which are FWD vehicles.

Even with data from all four wheels the system as it is described cannot prevent drift altogether. Wheel diameter is variable depending on many factors, tires can lose grip on the road, and also the earth is not perfectly flat. There would be other sources of inaccuracies, but basically in any real-world situation you need some kind of absolute position source to get drift-free location data. The trick is that it can be very noisy and low bandwidth if you integrate it with other sources of data like and IMU.

The comment you replied to was an attempt at a joke

This thread just keeps drifting further and further away from the original topic.

I know explaining a joke ruins it but I failed to understand it so far.

"Drift" is another term for understeer which is when the rear-end of the car loses traction in a turn. It tends to be easier to induce (accidentally or on purpose) in a rear-wheel-drive car.

I should note that my ~2005 MY Ford Focus would absolutely understeer if you enter a turn too fast without applying power.

Oversteer is when the rear end loses traction. Understeer is when the front end loses traction.

holy shit I wrote it exactly backwards. I didn't have my coffee this morning.

Are you lifting right before you start to turn in? That'll do it, because lifting off the throttle abruptly will shift weight to the front, which will unweight the rear; and sometimes that little weight transfer is all it takes to break the back end out a bit. (Used to do this regularly in a Taurus that I had the pleasure of driving for a few months.)

Yeah, the first gen focus is pretty easy to induce lift-off oversteer in. The rear end weighs nothing and the suspension sits strangely high.

> Lift-off oversteer

was on the tip of my tongue. Thank you!

They are intentionally misinterpreting "drift" as the driving technique (often associated with RWD vehicles) rather than the intended meaning.

The device is described as using a gyroscope to measure movement, so it would be independent of wheel diameter and grip. Gyroscopes still drift, so occasional calibration against an absolute position source is still required.

The device uses a gyroscope to determine direction. But a gyroscope is unable to measure distance travelled. So this device also attaches to the transmission to determine distance.

> A special servo gear was also attached to the transmission housing to feed information to the Gyro-Cator to help maintain position, map speed and distance traveled.

You could compensate for drift if you use a map as a constraint on the position.

You can drift with anything that has wheels, or even, tracks, somebody did that.