back to article Biker nerfed by robo Chevy in San Francisco now lobs sueball at GM

A motorcyclist is suing General Motors in the US after he was knocked off his motorbike by one of the automaker's self-driving cars. In a civil lawsuit [PDF] alleging negligence and demanding damages, photographer Oscar Nilsson claimed he was hit by an autonomous Chevrolet Bolt that had a human driver at the wheel. The driver …

Bronze badge

Re: I think the cyclist is at fault#2

"Drive with the presumption that every other road user is an idiot."

Likely the part dropped to dumb it down enough for the average road user is:

"and every one of them is looking for confirming evidence that you are an idiot, too".

1
0
Silver badge

> It is only USA which has the grand idiocy of everyone driving in any lane at any speed they please and overtaking on both sides.

No. That statement is not true at all.

It is not "only USA". Other countries, such as the that I live in, does not ban passing in any lane where they are marked. This does not lead to any particular problems.

Most countries, especially the USA, certainly do not allow 'driving at any speed they please'.

1
0
Silver badge

Re: @kain preacher

"but are you sure there is a law stating you must get out of the way of someone breaking the speed limit?"

Nothing that says that explicitly, but if the inside lane (first lane, whatever) is free then you should be in it even if you are breaking the speed limit yourself.

It's not the job of members of the public to enforce the speed limit, this only causes anger and makes things worse. It's common sense not to block the road for the sake of being a prig.

2
0
Silver badge

At JMCH...

That's the rub: clearance. If a vehicle starts to change lanes but then has to abort & return to their original spot, another driver anticipating the vacancy & proactively moving to occupy that (not yet vacant) spot has failed to leave proper clearance.

It really doesn't matter if either is a motorcycle or an autonomous vehicle, it only matters that the driver that did the anticipate/proactive bit failed to give the proper clearance for safe driving. The autonomous one probably would never do such a thing (it would detect that the lane isn't empty yet & thus refuse to move into it) but cyclists (filter/split) it all the time.

As my defensive driving instructors always drilled into my head, "You & your bike might weigh a thousand pounds. That car alone weighs a ton or more. Who do you think wins in that contest? Here's a hint: it. ain't. you."

I've often done the filter/split maneuver, but *only when I'm sure it's safe to do so*. Attempting it while the car I'm about to drive beside hasn't finished lane changes is outright suicidal; I don't know if the driver will change their mind, swerve back in, & turn me into a pap smear.

Yes a Human driver aborting the maneuver *might* look back to make sure their lane is clear for them to abort into, but they shouldn't have to. If they haven't completely left it yet then any other driver trying to occupy that same space would be guilty of trying to sideswipe me.

I agree you need to look out for number 1, but if you think that was a good time to try & pass another driver then you'll be reduced to a bunch of number 2.

=-J

5
0
Stop

Re: @kain preacher

In South Africa; if you didn’t move out of the way; you used to get done for ‘Depriving the State of Income’ and get hit for at least DOUBLE the speeding fine they thought they would have got if you hadn’t slowed the other car up !

1
0
WTF?

Re: I think the cyclist is at fault.

Especially when you watch cars ahead manoeuvring to ‘close the gate’ - sometimes quite late & aggressively; so that the poor sod on the bike suddenly as nowhere to go..

On the other hand; why are there no ‘Bikers THINK Car’ (maybe the ‘Car’ is an overstatement). Don’t bloody well hide in my blind spot(s) - or tail gate me; etc etc - I suspect that the car drivers who ‘close the gate’ are just pee’d off with some of the more stupid bikers

1
0

Re: I think the cyclist is at fault#2

If Bikers would only think - IF you can not see the drivers eyes in his mirror then he probably can not see you - get out of his blind spot

And if he is a dick & hasn’t adjusted his mirrors so you can’t find that sweet spot; then you know he never uses his mirror anyway - so now you know & can take suitable avoiding action.

2
0
Silver badge

"It is only USA which has the grand idiocy of everyone driving in any lane at any speed they please and overtaking on both sides. "

Perfectly legal to overtake on either side in NZ as well where there is more than one marked lane - the crucial point being that the lanes must be marked as separate lanes, not just a single lane that is wide enough to squeeze through on the left.

1
0
Silver badge

"Failing to yield to faster traffic is illegal"

The UK could do with a law like that. (Although it's legal to pass on the "slow" side, it's risky.)

0
0
Silver badge

Re: I think the cyclist is at fault.

"Lane splitting is already a dangerous thing to do even in the best of conditions & at the best of times"

Pretty much THIS.

There's a name for bikers who lane split in moving traffic - "Organ donor"

(in stationary traffic it's a little different, but never do it with a speed difference that doesn't allow you to abort - for exactly the reason this guy ended up off his wheels.)

1
0
Silver badge

Re: @kain preacher

"That's not true. In some states it is illegal not to move over to the right to allow faster traffic by even if you are doing the speed limit or greater."

In some countries the rule is "Use the left/right lane unless passing"

Failing to keep left/right is a ticketable offence in many countries, but one that's seldom enforced.

1
0
Silver badge

Re: I think the cyclist is at fault#2

"Just work on the basis that anyone using the road around you can and will, and entirely without warning, do something stupid"

And if you're a biker/cyclist, work on the basis that everyone is plotting to kill you and just waiting for the chance to do it. Being in the right won't hurt any less if you go under a bus.

2
0

From the New Zealand Road Code:

"You can only pass on the left when:

"there are two or more lanes on your side of the centre line and you are able to pass safely by using the left-hand lane

"you are directed to by a police officer

"the vehicle you are passing:

"has stopped, or

"is signalling a right turn, or

"is turning right."

So, "undertaking" on a motorway is legal in drive-on-the-left NZ (but they are thinking of stopping that).

1
0
Silver badge

Re: @kain preacher

Only the US would have a lunatic law that made it an offence to not give way to a breaking the speed limit (non emergency services) vehicle

0
0
Silver badge

Re: @kain preacher

tiggity, the US doesn't. Induhvidual states might, though.

0
0
vir
Silver badge

The Real Moral of The Story

Always count on the minivan to slam on the brakes for no reason.

18
0
Bronze badge
Facepalm

Re: The Real Moral of The Story

"Always count on the minivan to slam on the brakes for no reason."

The much-misunderstood principle "stop and think" in action.

1
0

If my memory serves me correctly, IF the biker was lane-splitting, then GM's account is totally legit because lane-splitting is only legal if it's safe. Otherwise the biker is in the right.

5
0
Silver badge

"IF the biker was lane-splitting, then GM's account is totally legit because lane-splitting is only legal if it's safe."

There's a circular argument at play here. If the car had completed its initial manoeuvre it would have been safe. It could be (and was) made unsafe by an action outside the motorcyclist's control. That could describe just about any movement, including continued movement forward within a lane, in heavy traffic.

I believe that in this case the car driver attempted to take over but was too late. If that's so then he thought that the collision was avoidable. And yet we're assured the autonomous vehicle will be so much safer.

0
0
Meh

Here in America most everyone is on the lookout how to make a buck the easy way. If the driver sues GM he is hoping that GM would rather settle out of court for a few thousands of dollars rather than the cost of many times that amount for them to take it to court even if they win. So take this with a grain of salt it will make this crap taste better.

12
2
Silver badge

"Here in America most everyone is on the lookout how to make a buck the easy way."

I was once told by a sales guy that the money in the other guys wallet is yours, he just doesn't know it yet. Capitalism at it's most ugly.

1
0
Silver badge

I've always just assumed that in the US it's easier to sue someone than it is to take responsibility for your own actions.

It's certainly more common.

2
0
Silver badge

" If the driver sues GM he is hoping that GM would rather settle out of court "

That's a long bet.

GM (or any other robocar developer) will fight this tooth and nail. The Last thing they want is any kind of legal precedent - even an out of court settlement tips the scales.

It doesn't matter if the biker was lane splitting or in the lane. He moved prematurely into a gap that wasn't available yet and got twatted as a result. The human driver trying to intervene _too late_ shows that a meatsack wouldn't have done any better.

0
0
FAIL

Regardless of who's legally in the right or wrong, a self driving car hitting another vehicle is a problem. Just because one party has right of way they should still be doing everything possible to avoid a collision with someone in the wrong - saying "I have right of way" and then ploughing into the side of another vehicle isn't clever.

No matter how this one turns out I hope GM are looking closely at their code...

13
1

My understanding is that "right of way" has to do more with who can proceed in a conflict, but loses value when a collision occurs.

Consider a vehicle that slows to make an illegal u-turn and is struck from behind -- Both drivers are fault, one for impeding the lane for an illegal maneuver and the other for failing to stop.

No "right of way" entitles one vehicle to impact another if the impact can be avoided, if either driver misses the opportunity to avoid a collision then they bare some fault which is why so many vehicle collisions end in "50/50" fault decisions.

5
0
Silver badge

"Regardless of who's legally in the right or wrong, a self driving car hitting another vehicle is a problem."

That was my point exactly. And immediately it turns into an argument of what's illegal in whatever state - about which, as a UK driver I know nothing.

If the car doesn't make best efforts to avoid an accident and its maker thinks that's OK because the other driver was in the wrong then we can forget about all those arguments about how they're safer and their proponents can stop going on about "meatsacks", gain some respect and start addressing drivers politely.

8
1

RE: My understanding

"Consider a vehicle that slows to make an illegal u-turn and is struck from behind -- Both drivers are fault,..."

Not in the four US states I have lived in, the car behind is (almost) always 100% at fault for any rear-end collision. No 50/50 for liability unless you could not reasonably have avoided the collision - like black ice or no brake lights in the front car. The U-turn car is also separately guilty of a traffic violation (and might get cited for that).

But, contrary to claims from some rather confused commentards here, US traffic laws do vary somewhat by state (in the state I currently live in you can legally pass on the right, but only on freeways or if the car being passed is making a left turn), and it is possible what you say is true in some states.

That said, I'll guess if you can make a legal left turn on that road, then your claim of contributory negligence (that the car in front is legally partially at fault for being rear-ended) won't fly in most of the US, as any car behind should reasonably expect a car in front of them might slow down to turn regardless of the reasons for that turn.

3
0
Silver badge

"Regardless of who's legally in the right or wrong, a self driving car hitting another vehicle is a problem."

Should read:

""Regardless of who's legally in the right or wrong, a self driving car hitting another vehicle is a problem."

1
0
Silver badge

“My understanding is that "right of way" has to do more with who can proceed in a conflict, but loses value when a collision occurs.”

NO

Right of way is who has the legal right pass over a specific place.

You are thinking of priority. If people stopped calling it a ‘right’ and called it priority then we might see a slightly more rational attitude on the roads...

Ah, who am I kidding - might is right, bigger vehicle always has priority in the eyes of its operator and anyone vulnerable is screwed by the lack of enforcement, prosecution and penalty :(

4
1
Silver badge

self driving car hitting another vehicle is a problem.

I don't agree on this. This may well become a test case for the manufacturers. If GM pursues it and wins, then it will set a precedent. Assuming they have sufficient telemetry I think they'll happily go to court over this.

Developers must continue teaching cars to drive defensively but, as the numerous "what if…" questions demonstrate, you cannot anticipate and prepare for every possible eventuality and there will, inevitably, be accidents. In such situations the law must be the same for people and computer drivers.

Insurance companies and legislators will be watching this carefully.

2
1
Silver badge

"Assuming they have sufficient telemetry I think they'll happily go to court over this."

Being a test vehicle in the R&D programme, they must surely have full 360o video, accurately time stamped. Even as a production car, I'd expect it to have a full sensor record of everything around it. All around proximity detection in an AV car is a pre-requisite so the telemetry should show the precise sequence of events.

1
0

"Regardless of who's legally in the right or wrong, a self driving car hitting another vehicle is a problem."

Problem is, in this case, the self driving car ended up with 2 choices - 1. drive into the minivan that slowed in the other lane, completing the lane change maneuver, or 2. return to its original spot, which someone had illegally ridden into.

From any point of view I can tell, this seems clear cut - the software did the right thing... You can't always prevent an accident, but you can choose which accident to have.

2
0
Silver badge

"Regardless of who's legally in the right or wrong, a self driving car hitting another vehicle is a problem."

The extra twist with a self driving car is that it's repeatable.

1
0
Silver badge

@Charlie Clark

There are two separate issues here. One is the the legal liability for the particular accident. The other is the implication that the self-driving car may have failed to take sufficient action to avoid a collision. GM might be pleased from a liability point of view if they win on the first. If that leaves them pleased with the outcome of the second then it's bad news for all of us.

2
0
Silver badge

"Problem is, in this case, the self driving car ended up with 2 choices"

Is this the case? Did it have an option to move over a bit, straddle both lanes and leave room for both the minivan and the bike? Did the car's S/W overvalue being completely within a lane? Did it have the option to brake within the minivan lane and avoid a collision there? Did the S/W overvalue the lane that it was leaving relative to the lane it was entering? Or steering over braking?

I've said here, on this topic, that accidents are the results of corner cases and that humans are usually better at handling those than computers. Was this a corner case that a human driver would have handled better than the car?

2
0
Silver badge

> Was this a corner case that a human driver would have handled better than the car?

Probably not - most humans would swerve 'back' across their lane if their overtake had to be aborted (assuming they didn't just plough into the minivan and spin through the m/cyclist).

1
0
Silver badge

"accidents are the results of corner cases"

Genuine accidents perhaps.

Most road crashes are the result of at least 2 major errors of judgement (usually 3 or more) on the part of one or more people.

The vast majority of crashes I've seen were avoidable in the first instance.

It's unlikely a human would have coped any better. All things being equal and the laws of phsyics being thwt they are, it's usually better to sideswipe something and exchange paint than than to run into the back of something else and suffer an abrupt change of velocity for both vehicles (Did I hear you say "whiplash claim"?)

1
0

All the "code" in the world, cannot prevent some idiot from driving into you. Dash cams.

1
0

Are self driving cars deaf?

I assume there is video of the accident?

The Bolt started to change lanes, then decided it didn't want to, so went back. One question is at what point is your lane change maneuver complete? Conservatively I think it would be an easy argument so that once both sides of tires are over the lane marker you have changed lanes, but one might argue that once you have mostly cleared the lane marker (maybe 50%?) you've fully committed to the lane change and to veer back over would be yet another lane change.

The more important thing though is, why did the Bolt feel it was OK to head back to center? Shouldn't it be aware of all objects around it and their vectors? This bike was accelerating into the space the Bolt wanted to be in, the Bolt should have recognized that.

...a third consideration--We've probably all heard the excuse that "loud pipes" on motorcycles save lives by making drivers VERY aware of the bikes nearby and based on the pitch/loudness you can tell if they are speeding up quickly, letting off accelerators, etc...

...it raises the question to me of whether self driving cars are "deaf" or not? I know that I certainly become more attentive when I hear the doppler-effect modified sound of a motorcycle racing up behind me... does the Bolt have this same capability?

7
0

This post has been deleted by its author

Silver badge

The car will have video of the incident

So if the motorcyclist tried to split lanes past the car before it had completely left the center lane, then he's at fault. Gonna be a bit harder to win these sorts of cases against autonomous cars since they will always have a video record from multiple angles.

8
1
Silver badge

Re: The car will have video of the incident

Lane spiting should not be done at more than 25 mph and should only be done if traffic is slow moving, close to a stop.

2
5
Silver badge

Re: The car will have video of the incident

"So if the motorcyclist tried to split lanes past the car before it had completely left the center lane, then he's at fault."

And are you then saying that the car was in the right to continue regaining its lane and side-swipe him?

4
2
Silver badge

Re: The car will have video of the incident

Absolutely, because the car never needed to "regain" its lane, it hadn't fully moved out of the lane. The motorcyclist was impatient, and his injuries were 100% his own damn fault.

7
5

Re: The car will have video of the incident

According to the article, the collision took place at 17 MPH.

1
0
Silver badge

Re: The car will have video of the incident

And are you then saying that the car was in the right to continue regaining its lane and side-swipe him?

No, the conclusion is that the motorcyclist acted dangerously and tried to overtake before it was safe to do so. In this interpretation it was the bike that did the swiping. But IANAL and would expect this go to trial to get a precedent juddgement.

1
0
Silver badge

Question?

Was GM car still partially in the original lane? If so, the motorcyclist is a candidate for honorary mention for a Darwin Award. Do not pass a vehicle until it has finished its lane change.

6
0
Bronze badge

I hope that General Motors will be defending itself robustly against these malicious lies.

0
8
Silver badge
Facepalm

Hmmm

If the biker was lane-splitting on the line between "fast" and centre lanes, the GM car would have had to have completed its manoeuvre, or the bike would have gone into the back of it.

And there's a difference between overtaking on the wrong side, and maintaining a steady speed when traffic in the overtaking lane slows down, else an accident in the overtaking lane would mean the other lanes had to stop. I'll happily sit at 60 in the "slow" lane with the trucks while the centre and "fast" lanes go 0->90->0->90->0

Closing the gap between the bike and next vehicle ahead would also seem reasonable, even if the "fast" lane slows down.

2
0
Silver badge

Re: Hmmm - lane splitting

Late to the party, but if the motorbike had been lane splitting then the car should have been able to regain the lane without impeding the bike.

That is, after all, what lane splitting is about.

1
0

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Forums

Biting the hand that feeds IT © 1998–2018