Hacker News Reader

Comment by Jensson | Hacker News Reader

> On Oct. 28, a Pony.ai vehicle operating in autonomous mode hit a road center divider and a traffic sign in Fremont after turning right

This is really bad, if their software can't even avoid hitting a stationary object then it isn't ready to test on public roads.

View on HN

> This is really bad, if their software can't even avoid hitting a stationary object then it isn't ready to test on public roads.

Almost as bad as a Tesla crashing into a stopped fire truck. Why is Tesla “autopilot” still allowed on Californian roads?

https://www.wired.com/story/tesla-autopilot-why-crash-radar/

Well, it wasn't driverless. The driver was arrested for DUI. Its not even known in that article that autopilot was turned on.

The article goes on to say that two other drivers who said they had autopilot on also crashed into fire trucks… How many fire trucks do Teslas need to crash into for it to be considered a problem?

One could argue that normal cruise control on normal cars would also crash into fire trucks. Ultimately drivers are responsible in both cases, it's their fault 100% end of story.

PS Tesla rolled a software update to detect emergency vehicles and slow down.

The cruise control on my Subaru probably wouldn't hit a fire truck since it has pre-collision braking.. but in any case, it's not sold as "Autopilot" or "Full Self Driving" so the limitations are pretty clear. There's a reason several countries stopped Tesla from marketing their nonsense like they do in the US - it turns out when you say things are self driving, consumers believe you and don't read the fine print.

https://www.reuters.com/article/us-tesla-autopilot-germany/g...

Airplanes on autopilot don't brake, either, when objects appear in their path. They especially don't in the case of fire trucks, or your Subaru.

Tesla's Full Self Driving has been released only to a tiny number of beta testers.

1) Most consumers have absolutely no idea how airplane autopilot works. Pulling a “well actually” on this does not remove the issue of misleading marketing.

2) Yes, and the videos I’ve seen from those beta testers are terrifying.

I actually think most people believe a pilot is needed to fly a plane with autopilot. Don't you? Who would want both of their pilots to be asleep?

No, I don’t think people believe that at all. A certain meme entered pop culture years ago that either the pilot is there for takeoff and landing, or the pilot is there in order to make the passengers feel safer. I certainly recall a lot of “planes these days basically fly themselves” articles. Neither of these are strictly correct about how autopilot works, but we are arguing about cultural understanding here.

Even taking the lesser of these two scenarios, this would imply that Tesla autopilot should be capable of doing most of the driving, with the driver just there for whatever bits you might argue are equivalent to a takeoff and landing. Given the stories of Teslas accelerating into stationary objects or totally losing track of the road lane, I don’t think even this limited understanding of what “autopilot” means is correct.

Direct quotes from Elon and the Tesla website:

> "The person in the driver's seat is only there for legal reasons. He is not doing anything. The car is driving itself."

> “Full Self-Driving Capability All new Tesla cars have the hardware needed in the future for full self-driving in almost all circumstances. The system is designed to be able to conduct short and long distance trips with no action required by the person in the driver’s seat. All you will need to do is get in and tell your car where to go. If you don’t say anything, the car will look at your calendar and take you there as the assumed destination or just home if nothing is on the calendar. Your Tesla will figure out the optimal route, navigate urban streets (even without lane markings), manage complex intersections with traffic lights, stop signs and roundabouts, and handle densely packed freeways with cars moving at high speed. When you arrive at your destination, simply step out at the entrance and your car will enter park seek mode, automatically search for a spot and park itself. A tap on your phone summons it back to you. The future use of these features without supervision is dependent on achieving reliability far in excess of human drivers as demonstrated by billions of miles of experience, as well as regulatory approval, which may take longer in some jurisdictions. As these self-driving capabilities are introduced, your car will be continuously upgraded through over-the-air software updates.”

Yeah, Musk blatantly lying about these things really doesn’t help their case one bit. It’s hard to hide behind “autopilot is just marketing” when their popular CEO is lying about its capabilities left and right.

If an airplane on autopilot encounters a Subaru, there are _much bigger_ problems than ‘not braking’.

It is important to note that suddenly, and against all probability, a sperm whale had been called into existence, several miles above the surface of an alien planet. But since this is not a naturally tenable position for a whale, this innocent creature had very little time to come to terms with its identity.

Most TACC systems won’t brake for objects going less than vehicle_speed-threshold, due to how radar works. You don’t get relative speeds and distance for a set of objects, you get a spectrum showing how much of what the radar sees is moving at each relative speed.

That’s also why most TACC doesn’t work below 30km/h or so. It can’t tell the difference between the car in front and a stationary object next to the road.

Edit: This is actually an interesting rabbit hole to go down, because on the face of it, it seems like a fairly straightforward system to implement. Then the more you think about how a 'simple' system should handle various scenarios, the more you realise that to actually do it properly, 'just keep a good distance from the car in front' requires most of the perception required for general free-form driving.

This isn't true (on Subaru at least, likely due to their binocular eyesight cameras...) The primary purpose of their suite of driver aides is to prevent accidents, so it's actually really good at that. There's plenty of videos to attest to that but here's an example of a 60kph stop when the system detects a cone:

https://youtu.be/0A4i4-xALVg?t=252

Edit;

I guess you did say braking with TACC is complicated -- but I think that's a bit of an edge case that only applies to Tesla. The Subaru version of TACC is ~fine but it never gives you the impression it will handle 100% of the driving for you, so you'd never leave it to navigate around emergency vehicles by itself. Maybe the solution to rarely dangerous self-driving is to make them only ~75% good so driver's are always aware?

That's pretty cool! Much better than the radar based TACC on the Hilux I rented a couple of weeks ago. I came away from that one with a strong impression that it was inconsistent in a scary way and shouldn't have shipped like that.

Do you have one (or experience driving one)? If so:

- How does it go with phantom braking and various objects on the side of the road?

- Does it distinguish between various on-road things? (eg. does it brake for a dog? a plastic bag? a puff of smoke? an anvil falling off the back of the truck in front?)

- How good's its understanding of lanes (eg. a very slow car that's in another lane but currently directly in front of you due to the road curving)?

- Does it brake for objects outside the lane entering it (cross traffic, lane cut-ins, pedestrians, vegetation moving around in high winds)?

By the same line of reasoning, you could say that my car doesn't have an emergency oxygen mask, therefore airliners don't need them either. If the airplane goes too high for passengers to breathe comfortably, it's 100% the pilot's fault, end of story.

Cruise control is literally just for (at least in a traditional sense) keeping your speed constant.

You're still responsible for watching traffic and using the brake or gas as needed, especially if there is some issue up ahead that requires a safety maneuver of any kind.

i.e. an emergency vehicle, like a California Highway Patrol vehicle, begins a traffic break 15 car lengths ahead of you. As long as you're observant, you'll notice that in time and either cancel the cruise control or make the right adjustment so you slow down enough to the break speed.

“Tesla AI detects imminent collision, optimises response time from fire department.”

The pre-collision avoidance should have kicked in either way. My Subaru has saved me once or twice when a car in front of me braked harder than I realized, and they’re not even claiming anything close to full self driving.

The Subaru wouldn't stop for a stationary vehicle though. Most of those systems either don't work at all for stationary vehicles, or work far less reliably. They're mostly designed for the case you described - car ahead slows suddenly and the driver isn't paying attention.

> Most of those systems either don't work at all for stationary vehicle

Citation needed. Subaru claims their eye sight system will detect and attempt to brake for a stationary vehicle. Whether or not they’ll managed to halt the vehicle is obviously speed dependent, but obviously slowing down before the crash is vastly superior to maintaining speed.

Care to clarify what you mean here? Had a subaru outback with eyesight a few years ago and I tested it with stationary objects in path multiple times (cardboard boxes and such). The subaru braked and stopped short of the objects 100% of the time.

When the vehicle you're about to collide with is moving slowly, there are good radar returns, so you can use both vision and radar to make the decison.

When the vehicle is stationary, the radar data has to be heavily filtered because radar has poor angular resolution and there are a lot of other stationary objects, many that you don't want to stop for at similar angles, like overhead road signs. That means you're relying purely on vision, which alone has a lot of false positives and negatives, so detection thresholds need to be set much lower.

Doesn't mean it won't work. Just means it's less likely to work, especially in the dark and the rain, where radar would normally be a big part of the signal.

When I was driving a 2014 Volvo, it also automatically stopped before hitting a stationary cardboard box (up to 30 kph)

Not all these systems are radar. Subaru uses cameras, in fact.

My 2015 Model S does that too. And also often detects stationary vehicles that it thinks might be in the way.

A Tesla driver was killed a few years ago in 101 south in Mountain View by hitting the central part of the fork.

https://www.engadget.com/2018-03-24-tesla-model-x-driver-die...

That particular spot had been crashed into before and the crash buffers hadn't been properly replaced. This led to a normal crash turning into a fatality.

The driver of the tesla had complained in the past of the tesla swerving into the wall when passing that spot, yet still had his guard down on the day he died.

That the tesla swerved is down to the poor striping on the road. The stripe along the right side of the left fork was brand new and very sharp, while the stripe along the left side of the right fork was extremely dim to non existant due to excessive wear sustained during construction of the left fork and the bridge it leads to (connection from the fast lane of 101s to the fast lane of 85s).

The crash barrier has since been properly replaced and the striping redone with diagonal lines hashing off the gore so it's abundantly clear even to a camera not to drive there.

That said, the tesla self driving made a very dumb error no human driver would make.

> That the tesla swerved is down to the poor striping on the road.

No, it's down to defective self driving.

If a self-driving mode depends on road markings being perfect to not kill me then I'd rather not have self-driving at all.

> That the tesla swerved is down to the poor striping on the road.

Certainly not. It's not appropriate for a self driving car to make any assumptions about striping or any road markings being correct. It crashed because it wasn't capable of self-driving after all.

>Why is Tesla “autopilot” still allowed on Californian roads?

Because Pony.ai is level 4 autonomy while Autopilot is level 2. The human driver is ultimately responsible for any accidents in a level 2 system. There is no driver to intercede or to blame in the case of a level 4 system. This crash is the sole responsibility of Pony.ai.

well they do have a "safety driver", they only removing pony.ai permit to drive without one.

>Why is Tesla “autopilot” still allowed on Californian roads?

Because Tesla suckered some idiots to take the blame if for example the car will instantly change direction and throw you of a bridge or in from of a truck, Tesla is saying that their customers are supposed to be ready for this cases .

For now only some Tesla alpha testers suckers got killed and no innocent person, but when an innocent will be killed I am waiting to see if the driver and hist lawyer will accept the blame or will fight the ToS claims.

I interviewed at a self driving car startup in 2016 that was testing their early prototypes on public roads! They were also using their investor money to rent out some ten million dollar home in the Silicon Valley hills as their office. Big red flags all around. I declined their request for a second round interview.

>They were also using their investor money to rent out some ten million dollar home in the Silicon Valley hills as their office

Is this supposed to be a bad thing? Why?

The focus on the price of the house seems pretty strange in an area with a particularly high price-to-rent ratio. But I guess $20k/mo just doesn’t sound as spicy.

I think he meant that they rent out their own home to their own company to extract investor money from the company.

That would be a very strange detail to leave out. But sure, it’s the only way their comment makes any sense.

Otherwise, spending investor money on office space seems like a perfectly normal thing to do. The $ amount described doesn’t seem too unusual for the area, and a house might offer significant advantages over traditional office space for a startup that is working on cars.

Anther explanation could be that many startups try to give out the vibe of "just look at how much investor cash we can burn, come join in on the fun!", and are very open about that. I can understand why you wouldn't want to join such a company, even though that is exactly what many others are looking for.

Sure, but now we’re just guessing. We could go on all day listing bad things founders could do, but the original comment simply doesn’t go beyond implying that it’s bad to use a $10M house in Silicon Valley as an office.

The price mentioned isn’t anything unusual even for fairly early stage startups in the Bay Area.

Well, when humans try to explain a culture that was alien to them they tend to use examples from that encounter. I can very well see the founder proudly saying "Look at this cool 10 million dollar office we have!", and then this guy latched on to that statement as being representative of their culture. When I work at an office I don't know what it costs, I doubt he would have looked it up if they didn't tell him, so this explanation seems likely to me. Of course to you that is not a good representation of the culture, so you protest here. Which is good, now we all know a bit more how to state things so others can understand it.

We're not guessing, at least not randomly, we're using intuition. Applying experiences and subtle cues to create something more educated than a random guess.

I see you have something like a half dozen comments now trying to rationalize this all cross the comment section, but like yeah renting out a mansion for your startup does send out a "vibe".

If you don't realize that a) you're probably who they're looking for, and b) you probably have some room to grow when it comes to picking up on subtext.

Not everything in life can be quantified in a neat little bullet pointed list. But part of being an experienced person is being able to pick out the more subtle aspects of situations.

To me a mansion as a startup HQ screams tech-bro 24-7 party culture with some major WLB issues.

And again, it's not like using a mansion magically causes that means that, but it strongly implies it.

It's like walking into an office and seeing everyone in suits and ties vs seeing everyone in sweatpants and baggy tees.

Technically the choice of clothes does not directly force you to behave a certain way, but your intuition should tell you that those are two very different cultures, and it should tell you a little bit about each.

tl;dr Ignoring intuition because it's not backed by unequivocal fact is not a super-power, and it's not really productive.

> If you don't realize that a) you're probably who they're looking for, and b) you probably have some room to grow when it comes to picking up on subtext.

Don’t get me wrong, I’d never work with the manchildren at comma.ai. I’ve interacted extensively with George Hotz in the past and he’s certainly not a person I’d want to spend any time in the same room with.

I just don’t see any issue with their choice of office space, it just seems like a perfect fit for their business.

> To me a mansion as a startup HQ screams tech-bro 24-7 party culture with some major WLB issues

Where else would you put your car tech startup in SF? Software engineers aren’t going to work out of a garage, and good luck finding a traditional office space with decent facilities for working on cars.

FWIW this is hardly a mansion https://www.redfin.com/CA/San-Francisco/41-Santa-Monica-Way-...

Comma.ai has a “proper” facility now, but they had to move off to San Diego for that.

> Where else would you put your car tech startup in SF? Software engineers aren’t going to work out of a garage

At the same time as the $10m mansion place I interviewed for a self driving truck company and they had their office located in an industrial park in a shop that looks like it used to be an auto mechanic with upstairs offices. That place was fine but I didn’t feel like working for a 22 year old CEO who had just left college. Back in 2016 investors were dumping money on self driving car startups but I wanted something stable.

> Where else would you put your car tech startup in SF? Software engineers aren’t going to work out of a garage, and good luck finding a traditional office space with decent facilities for working on cars.

In…a garage. I’m serious - there are spaces big enough in SOMA and other places that can fit a garage space plus open office space plus some conference rooms (I’ve worked in several of them).

To me it seemed more expensive than renting a boring office building with few workshop bays with roll top doors. So to me, they were burning unnecessary cash and this made it seem like they were inexperienced. The CEO was some university researcher who had done some self driving car work and then had investors throw I think more than $10m at him. I just got the sense that this guy was feeling flush with cash and wanted to burn some on a luxurious multi level mansion with a pool and a hot tub.

I had just gotten out of a startup where the guy treated the place like his personal clubhouse and had a really controlling and abusive attitude. I wanted to steer clear of any company that seemed inexperienced.

It apparently doesn’t look that way to you, but that’s how it looked to me and I see other people here can relate.

There are plenty of other reasons if you think through it even a little bit.

* Are the founders or employees living in their office?

* Are they treating it like a work environment or a party house? Are the work, kitchen, and bathroom environments professionally maintained.

* Is there adequate parking or public transportation?

* Is it legal? Does it conform to HOA and city requirements?

* Is it centrally located for commuting workers and testing sites or the personal preferences of the founders?

* Is it scalable for a growing company?

Renting a mansion as an office throws a hundred red flags to anyone wanting a serious outcome from a startup.

I've heard of tons of startups that do though, including some very successful ones.

Wasn't snap for example running out of a few houses on the la beach? Maybe they weren't mansions, but I'm sure the rent was high.

Most of these seem like separate issues. Founders living in the office is weird regardless of it being a “ten million dollar house”.

> Is it centrally located for commuting workers and testing sites or the personal preferences of the founders?

The house in question was probably the comma.ai office in West Portal, SF (right next to the West Portal Station). Draw your own conclusions regarding commutes, it’s a pretty central location.

> Is it scalable for a growing company?

Offices are rarely very scalable, you just move somewhere else once it no longer works.

It was not comma ai. The house was in the hills of Silicon Valley and not in a central location.

Seems like setting up in houses is a common pattern for Bay Area self driving car startups then. I’d guess that the availability of decent office space with workshop bays is essentially nonexistent

Not in the South Bay where this company was. Plenty of machine shops, mechanics, etc with workshop bays imo. What was true at the time was that investors were throwing money hand over fist at self driving car startups, even to young founders fresh out of college. I would not discount the psychological appeal to a 20-something founder with millions of dollars a to burn to set up shop in a mansion.

The point of my original comment is not to say that doing so is objectively bad, but I had just left an abusive startup run by a young and inexperienced startup and I wanted something more stable. I ended up getting a job at a Toyota research branch in a boring office park and I played with a cute little robot. It was just what I needed at the time.

You consider HOA requirements and single-site scalability to be red flags for a startup?

> * Are the founders or employees living in their office?

This is common for startups in the seed stage. I don't personally recommend it but I wouldn't consider it to be a red flag for a pre-series-A startup.

> * Is there adequate parking or public transportation?

There is practically no public transportation in the bay area. No, there is never enough parking.

> * Is it legal? Does it conform to HOA and city requirements?

Nobody takes a shit about HOAs. Startups exist to drill square pegs into circular holes. Hell, I'd love to create a startup to help people get rid of HOAs.

You get it. It felt like they were inexperienced and burning cash unnecessarily. I had just left a very unprofessional and in hindsight abusive startup and I didn’t want to walk in to another “boys clubhouse” situation.

No, I just meant that they were using some extravagant home in the hills as an office rather than doing the less exciting thing of renting an office in one of the many office parks. It felt like they were wasting money on a lavish lifestyle. To me it seemed like an indicator of a lack of prudent business sense.

I had also just left a startup where the CEO treated the company like some boys clubhouse and I did not want to get in to a situation like that again.

Not comma dot ai, was it?

The bit about big red flags all around seems about right for comma, but I don’t think they’ve ever had an office in SV.

Pretty sure in 2016 Comma.ai was at 41 Santa Monica Way in West Portal.

No it wasn’t them.

> This is really bad, if their software can't even avoid hitting a stationary object then it isn't ready to test on public roads.

Just slap a Beta tag on it and she'll be right.

That software bug will buff right out

They took "move fast and break things" a bit too literally I guess.

Move fast and break bones

If it does that more than people then it's not ready. People hit signs and dividers all the time. It's hard to come to realistic conclusions based on one case.

Depends on the details of course but I'd expect an autonomous vehicle to be much much better at not hitting things. Especially stationary objects.

People are bad drivers because they have limited vision, limited attention, are emotional and are easily distracted by all kinds of things.

Just because an autonomous car has a similarly bad track record as the average human, doesn't mean it's ready for the road.

It may be much better. This is a single case and we get every single case reported. The problem is that computer vision is not the same as human vision and at some point we may end up with a situation close to "this car drives itself perfectly apart from running over a completely obvious, well visible person once a year". And it's going to be a really hard one to process for many people.

A bad driver risks his own life so he has good reasons to try to improve. A bad self driving car program doesn't risk the founders life so he has no reason to improve it if regulations doesn't force him to.

It risks the founder's net worth so there are some reasons to improve, maybe not enough.

It risks the founder's net worth since we ban their cars from driving if they aren't safe. If we didn't then they would continue to run unsafe programs.

Quitting risks their net worth more, so they have an incentive to keep at it potentially regardless of results.

Is that same lever being applied to everyone thats testing this right now, or is this making an example?

I know nothing about nothing, so it's a real question.

Well, Tesla's autopilot has been slamming into emergency vehicles on the sides of highways for years and the feds have finally gotten around to investigating things.

Their full self driving beta requires constant vigilance because it likes to randomly veer at pedestrians and light poles.

As a sidenote, it's irritating that Reuters keeps calling it an "accident." It was a crash.

Every other self driving company out there has the same type of incident.

I think this one got revoked because the car was actually in autonomous mode without a safety driver in the car.

After they got a permit to drive without a safety driver? I don't think so. The permit should be for when the technology is mature enough to never do these kinds of mistakes.

Reply on news.ycombinator.com