←back to thread

32 points LinuxBender | 1 comments | | HN request time: 0.209s | source
Show context
alvah ◴[] No.42194679[source]
I very much doubt AEB saves anything like as many lives as "studies" claim, for the simple reason the "studies" will not be able to account for the number of times AEB hauls on the anchors, unexpectedly, for no reason at all. This has happened to me 3 times in the last few years, fortunately with nobody behind me, and likely causes more accidents than it avoids. It will be good when it's production ready.
replies(4): >>42194717 #>>42194787 #>>42194839 #>>42195256 #
HPsquared ◴[] No.42194787[source]
Lane keeping assistants can be pretty bad too. I have a relative with a 2018 VW that has a lane keeping function, and it's swerved the car out of nowhere a couple of times (or at least, maybe it made a small correction which startled the driver and caused them to swerve - either way it's a human interaction nightmare).

I think it's kept in some kind of standby mode at low speeds then kicks in automatically when over a certain speed, when IT chooses to kick in and can take the driver by surprise, the car tries to move to the centre of the lane (or what it thinks is the lane).

replies(3): >>42194977 #>>42195003 #>>42196151 #
vel0city ◴[] No.42196151[source]
> or at least, maybe it made a small correction which startled the driver and caused them to swerve

LKA is practically never strong enough itself to overpower an actual driver and force the car to swerve especially if the driver had even a reasonable grip on the wheel.

If the nudge of LKA was enough to startle the driver and have them swerve then just normal road feedback into the wheel would also cause them to startle and swerve.

replies(1): >>42196224 #
1. HPsquared ◴[] No.42196224[source]
Normal road feedback is predictable and has a central tendency. Who knows what the LKA will do when it does the equivalent of seeing a squirrel on an otherwise straight road.