Honda's now selling the first production car with level 3 self-driving
thedrive.com
thedrive.com
I'm assuming if it won't come to the US it won't come to the UK, Canada or Australia either.
Is 100 units really a "production car"? I don't agree.
Also:
"It does all this with zero input from the driver, who Honda says can "watch television/DVD on the navigation screen or operate the navigation system to search for a destination address."
and yet:
"Please do not overestimate the capabilities of each Honda Sensing Elite function and drive safely while paying constant attention to your surroundings. Please remain in condition where you can respond to the handover request issued by the system, and immediately resume driving upon the handover request."
[0] https://www.sae.org/binaries/content/gallery/cm/articles/pre...
[1] https://auto2xtech.com/volvo-to-skip-level-3-autonomous-mode...
The obvious answer is to apply AI to the entire system, not just the car navigating the road
* Is the driver paying attention, how much attention should they be paying verses the conditions?
* What is the environment like, how should I react, refuse to drive? Go slowly? Honk?
We already have self driving vehicles, they are called Mules. Why not make artificial mules instead of artificial pole position. The car should be an active participant, more of a copilot and than a blind automaton.
Self driving cars should be shown to have passed a repeatable rigorous adversarial gauntlet before being allowed on the road. And the individual cars should have to be re-certified every 6 months by retaking the test.
> When the feature requests, you must drive
Is there some kind of expected delay from the system requesting driver attention to the driver assuming control? Ie something like 3 seconds?
The more you "disconnect" as a driver, the less likely you are to take over in a split second and this has been proven again and again with humans. A system that works 90% of the time is probably the worst because it's good enough to give you confidence but bad enough that it's a false one. The driver is in the position to buy the advertised "you can disconnect" feature but then only be able to use it under explicit risk of harm to themselves and others. It actually increases the mental load as drivers will try to both "disconnect" (type on the phone, watch a movie) while also "paying attention to the road". Of course both can't be reasonably done at the same time so both experiences are sacrificed.
It's like encouraging people to take a taxi after drinking but then expecting them to take over whenever the driver makes a mistake and holding them responsible and accountable for the outcome. You won't enjoy the drink, or the drive.
These half-way solutions are great for marketing and for people who are all about the hype. But not only are they the bad compromises, they're also fueled by bad and incomplete data, and by marketing departments wanting to sell more. Almost no self driving outfit provides data on how many times did a human make the slightest correction while the self driving was enabled. Or out of all possible driving conditions, what percentage of them were covered by those self driving miles.
My personal philosophy is that cars are either self driving or they're not. Meaning they can either match an average human driver in all conditions they're expected to encounter over the lifetime, or they're just assisting the driver. If it's self driving except when it's not then it's just like a student driver. You wouldn't confuse them with being perfect drivers just because of their perfect safety record, given the supervisor corrects their every mistake.
The system gives you more power and you'll kill yourself. Similar situations with horse-drawn are not as lethal. As far as I remember, Volvo denied to use horseless carriages years ago, which is the only grown-up and responsible answer to this problem.
Edit: sp
Is this a Sesame Street reference?
Com...pjutaaa.
A friend who is an engineer at a premium car manufacturer working on sensors for self-driving is telling me that the internal policy inside the manufacturer is, that the only competitor that matters is Tesla, so they only compare themselves to Tesla. The same friend also believes that real self-driving is still many years out, even sensors that can deal with general weather conditions do not exist yet.
IMHO AI to analyse the data is what’s missing - we have cameras that are just as good as human eyes, so we know for a fact that the current sensors are enough to drive in general weather conditions.
You have two choices:
o have a sensor that gives you depth directly (ie laser, radar, or lenticular array) o try and infer depth from stereo cameras
Lidars are expensive and power hungry, Radar is cheap and mature, but doesn't give you anywhere near as much resoultion
lenitucular only has a tiny range. There are other time of flight sensors, but they are either not production ready, expensive or both.
What Tesla have chosen to do is kinda try and merge radar and monocular object detection to give a higher frame rate depth estimation of objects. However its expensive to develop, unreliable, and terrible in corner cases (ie, if sees an unknown object, it can't place its depth.) Now humans can do this, because we've had years of training, Tesla can't.
Tesla's fancy cruise control is dangerous. Its auto drive stuff its trying to develop is even more dangerous. Instead of finding the sensors and processing spec needed to drive safely, they are trying to use the sensors and GPU they already have. Its not going to work and is unsafe.
A camera feeding into phenomenally sophisticated AI trained on all sorts of weather conditions with windshield wiper smear and glare and everything else, which can turn those horrible images into a picture of what actually lies ahead, is an object sensor. Those aren't good enough yet.
However I agree that the mixed messaging is sad and irresponsible, and this article's author could have been more circumspect.
Because of that divide, I think there should at least be an online training curriculum and test before your car will unlock autopilot mode.
I actually find driver assist technologies almost as damaging. For example, lane assist (car centers itself in lane) can cause the car to veer when lanes are mis-painted. After having it turned on for a few minutes without issues, you begin to relax, then the car decides to take an off-ramp or swerve towards the shoulder.
As a comparative example, let's look at automatic gear boxes vs manual. As far as I know, there are no auto gear boxes that will occasionally require you to hit the clutch in order to successfully change gears. Their either completely auto, completely manual, or auto with manual override. Having something halfway between automatic and manual is just ASKING for problems; that being said, an auto gearbox that is expected to work only 99% of the time also really doesn't exist, which is why self driving is hard.
I'd much rather better alerting and imminent danger capabilities; more ubiquitous and accurate blind spot sensors, lane change cameras, brake alarms (or even auto-braking, if it's high reliability). These technologies will allow the overall self-driving landscape to improve (because there's so much overlap; self-driving NEEDS all of these sensing techs, and needs them to be REALLY good in order to work) and mature over time, while road infrastructure, mapping, legislature, and public opinion catch up.
That's very true, I've seen it as well. I think the big difference is that humans have the context and experience to deal with those situations on the fly, making choices that a self driving systems just haven't matured enough to make; this is why I advocate for better alerting systems rather than anything that changes the trajectory of the car on its own.
Self-driving: car comes with no steering wheel.
Lane/drive assist: car has steering wheel and you have to steer the whole time—with occasional nudges from the AI.
It'll lead to the same number of deaths and slightly more complicated lawsuits.
At a statistical level Nobody(TM) is heeding fine print warnings.
Is this true? I was just purchasing a car and did a lot of research. I was looking only at the European brands because I like the overall offering. There are different names, but none too flashy or promising too much.
How they call the driving assistant systems:
- BMW: Driving Assist (Regular/Plus/Professional)
- Volvo: Pilot assist
- Mercedes-Benz: they just describe the functions, i.e. active cruise control, steering assist, etc.
- VW group (VW, Skoda, Seat): Travel assist
Screw actual utility, this is about revenue, plain and simple.
We're not at a point where it's safe for drivers to move their attention from driving, but the technology is mature enough that it can and should intervene wherever possible to avoid collisions and other incidents.
The expectation that drivers can divert their attention from driving to perform other tasks on the expectation that they can resume control at short notice is extremely misguided. This should not prevent similar sensor and software tech that could enable L3 autonomy from enabling more effective accident avoidance technology that intervenes in situations where the driver may drive in a careless or dangerous way.
Teslas at least get better over time. If it got confused over a fork in the road, there's a decent chance that after a near-future firmware update that same car will no longer be confused. A few folks have acknowledged that they're now smart enough to slow down for curves (even sometimes a bit generously), and they are among the few that work pretty well off-highway (most of the time).
Words matter
Perhaps the use of this system should be licensed separately from a standard drivers license, akin to an IFR rating. And to keep the rating, manual driving must be performed periodically and logged as well.
If it turns out this is not an issue and we unnecessarily made self driving cars less common, that would have the same effect as shooting a few hundred random people every day (3,700 people die every day from crashes)
We'll find out in a couple of years
Can you provide your source please? Closest I could find is https://www.cdc.gov/injury/features/global-road-safety/index... which states that “[e]very day, almost 3,700 people are killed globally in crashes involving cars, buses, motorcycles, bicycles, trucks, or pedestrians. More than half of those killed are pedestrians, motorcyclists, or cyclists.” I didn’t find a number for just motor vehicle crashes, and since the rates are three times higher in developing countries, it seems like the cost of ownership is going to be a huge factor.
Although beating the average is arguably "good enough" to deploy in a collective sense, we live in a society of individual actors, and if I'm an above average driver, it needs to be better than me, not better than average, for me to want to use it.
Assuming a society in which each person individually and rationally chooses whether or not to use it, if you want 99% of people to use it, your software needs to be better than the 99th percentile driver.
You've got Jose the 35yo MRI service technician who logs 100k/yr for work and has had so much time to get good he can tell exactly what traffic is doing and are highly capable of printing what is about to happen and are preemptively making moves based on that. He knows his insurance company would crucify him if the saw how he flings an overloaded Transit Connect through an on-ramp or parallel parks by braille so if you press the issue he'll tell you he's good at getting where he's going but that the bureaucrats who write the state driver's manual wouldn't like him.
And then you've got Karen the elementary school teacher nearing retirement who logs 10k/yr 5k of which are spent looking at her phone. She has spent a cumulative one hour of her life above the speed limit despite spending much more than that on highways where the traffic flow is well above the speed limit. She follows every rule in the book to the letter, gets honked at daily and once a week she has a story about "some asshole" she got in a conflicting situation with. She doesn't know how far down the gas pedal on her 4Runner goes but oh boy does that brake pedal get a workout when "oh crap almost missed my turn". She swears up and down that she's a good driver because of the seventeen fender benders she's been in she was only at fault in the three that were caught on camera or where witnesses stopped.
Which definition of "good" do you want near you when there's 3" of snow on the road?
It actually can be true, depending on the distribution of skills you assume. Say you have 100 drivers, 99 with the same skill level, and one driver which is worse. 99% will be better than the average. I'll see myself out ;)
You should be happy that other people can buy them because they'll no longer crash into you ;)
But if a company sells a product with the name "Autonomous Driver", than that company is liable if the product mulfunctions (ie. drives a car into a tree).
If I was a car manufacturer and I knew that my log files would undoubtedly be used as evidence against me in a criminal negligence trial, I would think very hard about what I did and didn't put into those logs.
At least a partial liability will have to fall on the OEM that's why nobody is making any binding promises (including Tesla)
And I am not sure any level of AI has reached that level yet.
Human drivers do not have those advantage.
So Tesla is comparing drivers in situations where AP doesn't even try to drive the vehicle because it couldn't.
This is amazingly misleading. And Tesla knows it.
The most prominent member of the TSLAQ people and (IIRC) the first to publicize a spreadsheet of Tesla deaths is Elon Bachman (a pseudonym). He has also been a charlatan on COVID. A year ago he said, “After 4 months of white hot Coronavirus panic, the disaster is visible everywhere except in the data. Total deaths, rounded to the nearest percent, remain 0% of annual flu deaths, and serious cases are falling”[1]
I tried to bet him up to $1,000 that US COVID deaths would exceed 25,000 by the end of 2020.[2] He ignored me and continued to deny the harm caused by the disease.
One simply cannot trust anything he’s involved in.
1. https://twitter.com/ElonBachman/status/1237030292340314112
I presume that the aim of the site is the opposite though.
I've seen more then enough crashes and deaths in Telsas that were perfectly avoidable providing the driver wasn't watching their ipad.
I've seen emergency vehicles attending a crash perhaps once or twice a year but never seen the crash occur. In fact the only ones I have personal first hand experience of is the time I rear ended a car, the two times that I was rear ended, and the time I slid off the road on an icy corner. All three rear ending events were relatively low speed incidents (under 30 km/h) and would quite likely have been mitigated or even completely avoided if the cars in question had had automatic emergency braking. The icy corner was my own fault for not thinking ahead.
So, to me, your statement, without some more context, sounds like an exaggeration.
That's different.
I have raced go-karts, I've raced GT cars, there is absolutely no way to keep aware of your surroundings if your focus isn't constantly on the road and checking mirrors, full stop.
You are assuming “aware” means “ready to take over in an instant”. That is not what it means in this context.
As stated countless times before, autopilot in planes isn't even geared towards handing most flight scenarios or challenging conditions. Tesla is technically correct to call it this, even though this naming is confusing to consumers who think "plane flies itself" and think it means their car can drive itself in all conditions and avoid at-fault incidents - which most likely will indeed lead to more deaths.
As stated countless times in response, it doesn't matter what autopilot actually does when the concern is the public perception of the marketing.
If "autopilot" sounds to the general public like "it drives itself" (which is the simple etymology of the word), then it doesn't matter one flying fig what autopilot on planes actually does.
From a marketing perspective however, robotaxis are just around the corner.
But I have the same concerns about all marketing that positions cars as being capable enough that people don't have to pay attention. Until we hit true Level 4 or Level 5 self driving cards, I'm extremely concerned about the public's perception that they don't have to operate a motor vehicle. *Especially* if the vehicle is capable of enough automated driving that the driver doesn't need to pay attention during most of the operation of the vehicle.
To your specific point though, I think in general the public would view something marketed at "autopilot" as more capable than something marketed as "co-pilot", despite the significant capability advantage that a "co-pilot" actually offers over an "autopilot" inside the cockpit of an airplane.
It's another perception/reality thing. I realize that the co-pilot shares the load, and is capable of being the pilot-in-charge, or having control of the aircraft. But one implies self piloting (Tesla), the other implies an intelligent assistant pilot.
Co-pilot says that the two of you are doing it together. Co-operatively.
Auto = automatic = you no longer have to do it.
Think of an automatic vs. stick shift car. And automatic car you no longer have to shift.
What could save lives is intelligent automatic breaking - because that is something we are currently capable of. Humans as a species is terrible at paying attention to boring tasks and quickly react - so these gimmick autopilot takes are dangerous if anything.
If the average consumer was also an airline pilot then it would make sense to use an industry term with nuanced meaning. Instead, the average consumer assumes a straightforward meaning of the term, especially with pop culture throwing out phrases like "planes practically fly themselves these days".
Claiming "autopilot is technically correct" would only approach being true if Tesla & others went into great detail to educate buyers about the limits of the avionic autopilot features referenced by the term, and difficulties of flying a plane in non-optimal conditions, so they actually understood the "technical" meaning and could apply it correctly to the car they're buying.
It's not even flying the car!
https://jalopnik.com/a-tesla-stan-dmd-me-to-show-what-they-t...
Since Japan drives on the left side of the road[1], a car manufactured for the Japanese market would have its steering wheel on the opposite side of a US or Canadian vehicle. It would be OK for the UK or Australia, which also drive on the left, but the cars can only be leased for now[2], and the provisions of the lease would probably disallow taking the car to another country.
[1] https://en.wikipedia.org/wiki/Left-_and_right-hand_traffic#W...
[2] https://asia.nikkei.com/Business/Automobiles/Honda-launches-...
Jeep has a RHD Wrangler for 2021 and I’m not certain if they ever paused making them.
Yes, technically it's easier to bring this car to other RHD countries but I'm assuming Honda has plenty of manufacturing capacity to do that if they want to. It just seems like they don't want to.
Considering you couldn't even get into Group B rally homoligation with those numbers (200) in the 80s, or Dakar race spec production (2500) in 90-2000s I don't think it is, either.
> It’s possible they are selling these at a loss to seed or whet the market. If it takes off then maybe it will reappear at a profitable price.
Exactly. It seems more like a mass produced prototype, or a proper limited edition model and if I'm honest this looks exactly like an Accord/Insight with all the JDM goodies we don't get in the West. But this is about marketing.
I wish them well, Honda (auto) has a ton of muck in their face due to their failure to, once again, make any inroads in Formula 1 after their epic success with Mclaren in the 80s--this last time around was so hard to watch, and the mere utterance of 'GP2 engine' will live in infamy forever. Which it's raison d'être to trickle down it's production car, but with V6 turbos and KERS systems pretty much maxed they need something like this to justify their R/D budgets.
I'm not sure how to feel about it, to be honest: had this been on an limited edition EV insight it would definitely be a step in the right direction. Whereas this will fall into obscurity as more manufacturers move to EV.
Unless your name is Lancia and you move 100 cars from one parking spot to the other while you're treating the officials to a lunch with loooots of wine.
I really wish Clarkson would do more of these kind of stories, the trio would be the ideal team to make a docu-series for the Motorex and GTR fiasco!
It's such an insane story that I cannot believe they're on the 100th version of Fast and Furious but we have not seen this story be told.
Having lived through it myself, I was in the early drifting community in SoCal back in 2002 and a regular on Fresh Alloy/NICO since 2000. I even saw a few of those cars that got sold at the old meetups at Life Plaza before the canyon runs, the Bee*R rev limit kits on those R chassis [0] were so absurd back then when they first came out, they sounded like the meanest rally cars.
My memory is fuzzy after all these years, and I can't remember if it was big bird, or black bird but we saw it being tuned on the freeway in LA doing high speed wangan runs and the helicopters may or may not have showed up, good times.
I'm going to miss the ICE days because of this, but... they're dinosaurs and we can always tell the story from the glory days. I'm just glad I was born when I was, because I think that is a culture that has since peaked and is now on a very sad descent, other than Tesla and Rimac I don't see anyone even trying to make EV cars be anything more than a soulless utilitarian computer enclosure with wheels to get you to point A to B.
Isn't this different from things like Tesla's autopilot in that in certain conditions the Honda will take full responsibility for driving and the driver can concentrate on other things. But needs to be ready when the car tells the driver to take control/responsibility. I'm guessing these conditions could be quite specific and limited - like traffic jams and freeway driving when the car is effectively in a 'convoy'
What any sane engineer would do is to keep testing these cars with a safety driver and gather anough information/miles to claim higher statistical safety than a human driver, while keeping track that those miles were really in fact driven without intervention/(or with predictable and managed intervention). Once they would have those numbers, then they would know for sure that the car is safe and they could release/manufacture hundreds of thousands/unlimited number of units to the market. If they didn't - it means either the car really isn't tested or it doesn't do what they claim it does.
Releasing a 100 vechicles just doesn't make sense, unless the claims are somehow shady or misrepresentative.
Of course, if it _doesn't_ do what they say it does, they will, ah, also make a killing. Or many killings. You can understand the caution.
He also promised it by the end of 2018.
Oh, and before that, 2016.
Sounds more like a beta test.
To me, it appears to mean "don't sleep or 'drive' intoxicated." The car can presumably predict when it needs to handoff with enough notice to allow you to regain the full focus needed to drive.
This is the critical assumption. And:
> suggests an unpredictable situation.
All driving situations are unpredictable. From the kid running into the street after their ball to the drunk guy in a truck careening into the same crossroad.
We can never reach this dream of "watching a movie while the car drives" without railroading our roads in some form or agreeing on a new distribution of risk.
So where Tesla might want to put this sort of tech into the hands of as many customers as possible, as soon as possible, Honda is being Honda and entering the water very slowly.
"Summon your car while dealing with a fussy child."
( do not summon your vehicle while distracted, maintain constant visual contact with the vehicle)
"Full self driving capability"
" vehicle has all equipment capable to self drive, but may be limited by laws or regulations in your area"
and so on. Tesla is full of nudge nudge wink wink, and disclaimers that walk back their ledes, so what exactly is problematic about Honda's statement about limitations?
Considering the odds, it is likely you would see no accidents if you only look at just 100 Tesla cars.
This would be more impressive they put 100,000 cars on the road and then we saw the stats.
Not to mention that plenty of non-self driving cars run into stationary vehicles on both motorways and ordinary roads.
A fireman I spoke to says it happens all the time and that the fire engine is parked upstream of the incident that they are dealing with so that errant vehicles run into it rather than the fire crew.
It is a matter of lively debate in the UK right now with regard to 'Smart Motorways' which have no hard shoulder, several people have been killed in collisions with stationary vehicles.
In fact even on motorways with hard shoulders a number of people have died because they stopped on the hard shoulder because of a breakdown and another vehicle strayed onto the hard shoulder and rear ended them at 70 mph.
Much of urban Europe with pre-industrial-revolution cities and part of urban Asia is slow moving traffic. Slow, compared to California/Atlanta/Houston non-rush-hour freeways.
- probably easier to have a japan-only dataset
- probably easier to get engineers to work on this at headquarters
- japan highway markings might be more predictable than say 50 states all with subtly different rules and state of repair
- us liability law
The unavoidable penalty for this is a longer period of driver responsibility and perhaps never reaching level 3. It's a trade off tesla buyers seem willing to accept.
Well, OK, sure, it's a possibility, but the reason why that wont happen is because we as a nation are much more litigious and expect to have a right to sue. We love to blame big business and make them pay.
People have been clamoring for reform of medical liability for decades - still no progress on that front, either.
Safe harbours are not politically popular here, and you tend to get them only in very niche areas that are below the radar of most people. Auto accidents and traffic injury law, not so much.
Tesla "auto"pilot does more, while pretending to enforce an always alert driver. This has resulted in a couple wrongful deaths no one at Tesla has been prosecuted for.
Honda meanwhile is taking advantage of the Japanese governments legal changes which make level 3 legally tenable. Drivers need not watch the road, but they must respond when the car asks them to. This means Honda is promising your Honda will not silently drive you into a wall as the Tesla autopilot is known to do.
To be fair amongst automakers, not many other mass-market car is different (right now) - they only auto-steer if the user applies torque to the wheel periodically and beep at the driver if torque isn't applied every 10/30 seconds (30 seconds is only set for interstates from my experience). Only sometime this year will the Ford Mustang Mach-E get an optional $600 upgrade to enable hands-off-the-wheel driving which uses hardware (already installed) in the steering wheel to monitor the driver's road attentiveness[0].
0: https://www.slashgear.com/ford-active-drive-assist-mustang-m....
e: no other -> not many other - apparently Cadillac has eye tracking
It won't drive to a ditch if I let go, but will start howling after 20-30 seconds of not touching the wheel.
Over here, the speed limit with road works is 40km/h
Overall, the problem of course comes down to reaction speed. An attentive human with hands on the wheel might respond in 250ms. A computer obviously can respond faster. But some problems won't give you any more than that 250ms to respond. Situations where faster sensors or cpu are meaningless in giving the driver more warning because there is literally only 250ms from onset of stimulus to the time needed for a reaction if catastrophe is to be avoided. Meaning that if the computer can't respond without human intervention, an inattentive driver is screwed because their 250ms reaction speed was based on them already paying attention. To break it down in rough estimates:
1) The car detects & make its own assessment and raises an alarm. 1ms? 10ms?, It almost doesn't matter.
2) The human receives & registers the alert. This is the initial 250ms reaction time.
3) Only then can the real work begin: The human, from a cold start, needs to take in the entire situation and understand what's going on. This is more than just reaction speed. This is task switching, which is cognitively more costly, especially for un-practiced tasks, and the task here is rapid analysis & synthesis of a panic inducing situation that most people never experience. 750ms is probably about right, based on research that shows the cognitive burden of task switching can be even higher [0]
So, a full 1000ms, a full second.
This is way, way too slow. As I said at the beginning, there are moments driving where that 250ms is required to avoid catastrophe. When those situations arise then there is simply not enough time for the car to get help of an inattentive driver to respond. It's simply not possible. Either the car makes the decision, or the human has just enough time to go wide-eyed before it's over.
Basically the only situations where the car's notifications can facilitate meaningful human response are those that can be foreseen at least a full 1000ms in advance. Certainly this encompasses a lot of difficult scenarios, but not many of the most dangerous ones. Any time a split-second decision saved the life of a driver? That same driver, if inattentive when the car's alert comes in, is now dead.
There is simply no self-driving car that that can safely allow human inattention until the car can make a whole lot more of its own decisions.
In the meantime, if overzealous marketing & breathless hyperbole don't continue to oversell these capabilities, then computer-assisted driving by attentive drivers should greatly improve safety.
[0] https://www.semanticscholar.org/paper/Task-switch-costs-subs...
If Tesla only sold 100 vehicles, I assume it could also have avoided them driving into walls...
If Honda had a million cars on the road, I would not be surprised if the accident rate was well over that of even humans drivers.
It certainly wouldn’t be statistically significant but over the course of a year or two you’d except a couple accidents for 100 cars.
... as previous versions of Tesla autopilot have done.
Talking about any one of these systems via their branding as if they are a singular fixed thing (and not a network of software versions and models under continual update) leads to bad mental models.
Seems like when the Street View pic was taken this was not a lane. I'm shocked it was converted into one. Seems very dangerous, for human or AI.
But that is the stupidest way of temporarily? (I hope) relaning a highway ever. What were they thinking?
One of those situations where Musks insistence on not using LIDAR is clearly wrong.
I own a TM3 and I do not have the "FSD beta" that is in limited testing. I will say that the car display portion of the UI really does not impress me in what it shows and does not show.
As in it is damn happy to show cones and garbage cans but I have instances where it showed me cones front and back of a vehicle but did not draw the vehicle in between. Same with a floating garbage can where sometimes the loader was visible but the truck itself would wink in and out.
I really wish people would quit ascribing super natural abilities to LIDAR let alone superiority. Plus I really want to see studies done which show the impact having hundreds of LIDAR equipped vehicles would have on people, animals, and even insect, life. (then again how many vehicles in traffic does it take to cause an issue for them)
I fully agree with you that that spot on the road is pretty horrible, which is why I mentioned it. But dealing with crappy roads is the inherent problem any self driving car has. It’s the car’s job to not hit things.
I own a Model Y and I feel like Tesla’s autopilot gets a worse rap than it deserves. But there is an inherent disconnect between what the car is capable of and what some people presume it is capable of. Tesla doesn’t help this with their branding.
Why did the driver just let his car plow into the pillar instead of taking control? Were they even looking up at the time? Would you have let your car drive into a pillar like that? I’d like to think I would have, but perhaps not.
Here is another doing a burnout and crashing into a store, a month ago: https://m.youtube.com/watch?v=tJu5TZ6rwXg
Human-driven truck runs straight into a 100-car pile up, 1 month ago: https://m.youtube.com/watch?v=VqNy7v5YekM
Beat that, Tesla!
Teslas per-mile incident rate is currently about 1/10th that of human drivers.
https://asia.nikkei.com/Business/Automobiles/Honda-launches-...
> It can free drivers from driving in congested traffic on an expressway when travelling slower than 50 kilometers per hour.
> The system automatically accelerates, brakes and steers while monitoring the vehicle's surroundings, using data from high-definition mapping and external sensors.
Even a head-on collision (as we've seen with Teslas) with a concrete pylon should be survivable at those speeds.
It's a bit of an overstatement to say Honda is leading self-driving. This seems like a great regulatory win in Japan, but not a very interesting technical achievement.
- hands-free overtaking of slower vehicles when at highway speeds
- a "traffic jam mode" that is hands-off up to a certain speed, and plays a video on the center display
This seems... not that impressive? I haven't seen many real reviews yet though, just a bunch of press releases filmed on closed courses.
The fact you watch a video while it's active is a tiny difference to the driver (especially since some people are already watching videos using LKA), but a HUGE difference in terms of what the system is doing. The fact the system is driving with full responsibility on Honda
To the point that instead of seeing existing LKA as a step behind this, you might as well see them as existing on different planes, where progress towards A does not affect progress towards B meaningfully
Yeah. I guess we'll have and wait for real reviews to see under what conditions this operate (or more importantly, fail to operate and hand control back to the user)
Nonetheless, o, the hype! Some caveats:
1. 100 vehicles is not "production."
2. "Please ... drive safely while paying constant attention to your surroundings." is not Level 3 Automation. The point of Level 3 automation is explicitly that it should allow the driver to not pay attention to their surroundings under limited circumstances, while the vehicle fully does the driving. The human remains the fallback mode: they must stay awake and alert so that they can take over control if the vehicle requests it. However they are NOT required to maintain a constant supervisory function. If they are, then that's Level 2 automation.
The problem with Level 3 has always been that it requires the vendor to take liability for unsupervised autonomous driving (like levels 4 and 5), but only some of the time, and the driver to take liability the rest of the time. The handover from one to the other is fraught, with the humans almost certainly being the weak link, since they're not all THAT good at staying alert in the first place, and get much much worse when they don't have supervisory duties to keep them awake. For this reason, many vendors have (wisely, IMHO) chosen to skip Level 3 automation entirely.
So my synopsis of this would be:
1. Honda's engineers have come up with some genuinely interesting new capabilities for autonomous driving, albeit falling well short of what would be needed for full Level 5 autonomy.
2. Honda's management would like to convert those innovations into some good PR for the company.
3. Honda's lawyers, actuaries, and human-factors people have realised that there's no way in hell mass deployment of Level 3 automation is a good idea, and so to limit the potential damage, are only allowing 100 vehicles to be shipped, to pre-qualified and well-trained buyers, and then furthermore attaching disclaimers which knock this down to L2.
4. L5 Autonomy remains a long ways away.
This is not much compared to Tesla's 3 billion autopilot miles and Waymo's 20 million driven miles.
Quality of course matters too, but the more information you have in different environments, the better.
Tesla/Boring almost did it when they were considering having electric sleds moving cars through Boring tunnels.
If there's a schedule, pre-planned route, and a fenced off road, there's barely any need for fancy sensor suites & AI. The most difficult part of it is probably that the road then couldn't be used by all types of vehicles.
Level 2 means that the car automates several different parts of the driving task, but that you must be ready to take over at a moments notice - you still have to keep your hands on the wheel and your attention focused on the driving task in case you need to intervene. This is what Autopilot, and the FSD beta is. You can turn them on but you have to keep your hands on the wheel and your attention focused. This also means you're legally liable for any accidents, since the onus is still on the driver to intervene.
Level 3 basically means that, in certain situations, you can take your attention off driving completely and play on your phone, read a book, etc. The car might still need you to take over in certain situations, but it will warn you in advance and give you enough time to transition back to driving. Companies have largely shied away from pursuing level 3 system because it puts the legal burden on them - if the car gets in a crash when it's operating at 'level 3', the onus is on the company, since the driver was not told to be ready to intervene.
WRT your point about how they could 'surpass' Tesla, two things:
Level 3 can be pretty restrictive - I'm not sure on specifics, but they could offer this on, say, only the ten largest interstates in Japan, and it would still qualify as level 3 self driving. The 'levels' aren't really a strict progression in capabilities. A level 2, 3, and 4 solution is really three separate problems to solve, not succesively better versions of the same system. It's conceivable to me that a legacy automaker could focus on getting this working on a few major interstates and have a system they're confident in after a few years, since they don't have to worry about nearly as many edge cases as a general highway ADAS solution.
Tesla's data/fleet advantage is overstated in their messaging. Most of the stuff presented in their autonomy day presentation doesn't actually exist (follow some Tesla reverse engineers for more info). Self-driving/ADAS is not, by and large, a data problem.
It's widely accepted that the disengagement stats companies are required to report to government agencies are essentially meaningless. There's no real standardization in how any of this stuff is reported. There's no way to do independent verifiable data driven analysis of the systems.
Waymo's the only company that's put out a really robust deep dive into their safety stats - the white paper they out out was really pretty comprehensive.
Legal liability tells you how confident the manufacturer really is in its technology.
can you explain more?
level 1 = expect car to go into ditch if you're not careful
level 2 = hands on steering wheel but it should be ok (Tesla)
level 3 = hands off steering wheel, looking around, expect to have to take over control (Honda's beta test)
level 4 = either taking over control rarely needed, or vehicle has limited ability (e.g. can't drive in snow)
level 5 = you can sleep
The confusion lies around whether we're allowed to watch TV - however I'm fairly sure no country or state has passed legislation allowing us to take our eyes off the road while driving.
Level 3: driver is expected to take control, with brief notice, when vehicle cannot determine correct action.
https://www.nhtsa.gov/technology-innovation/automated-vehicl...
Level 2 requires instantaneous takeover should the system fail, Level 3 system must tolerate ~30s delay, etc. in feature checkbox style.
Automakers are fully committed to the Level system, but it does not agree with the reality that algorithms don’t fail gradually like a turbine engine would, so the reality do not agree with the Level system and Level 3 is just going nowhere.
Your confusion around Level 3 comes from the fact that Level 3 do not exist. So no worries.
I'm not confident L3 actually exists. The only real levels are 2 and 4.
It's pretty "easy", actually, by making the take-over optional. I.e. you need the driver to take over within a few seconds if you can't navigate, but you'll fall back to a safe stop if the driver can't take over. So your auto-pilot doesn't need to handle all situations (well), but you can safely be inattentive; it might just cause a sub-optimal response.
This changes the legal situation for car manufacturers dramatically and has technical implications. If the power in your car for the controlling device fails, normal cars have the driver as fallback to break. Honda needs to have some technical redundancy somehow.
Tesla acts as a third party. They leverage a tort system not designed to accommodate the type of telemetry they have to aggressively find fault with the drivers action.
https://www.sae.org/news/press-room/2018/12/sae-internationa...
TL;DR: In levels 0-2 you are driving the car and the electronics are helping to manipulate the controls. In levels 3-5, you are handing over the responsibility to drive to the car. You do not need to babysit the car to make sure it's not trying to do something stupid.
"Please do not overestimate the capabilities of each Honda Sensing Elite function and drive safely while paying constant attention to your surroundings. Please remain in condition where you can respond to the handover request issued by the system, and immediately resume driving upon the handover request.""
How can you immediately respond and pay constant attention if you are watching a DVD?
I have the 2019 Accord with the regular Honda Sensing and this happens in multiple situations. For example, the car will turn off automatic cruise control if the weather becomes too severe. Snow can also obscure the sensors in winter.
Some roads are also too damaged (no lines, no barriers) for the lane keep assist to work and you need to control the steering manually when this happen. But that is probably not an issue for the honda sensing elite.
Only 100 Honda Cars with Level 3 system, seems like a marketing gimmick.
Here is Toyota demo from May 2020 of a Tokyo metropolitan expressway, entering, lane change, and car overtake, with a reporter:
https://toyotatimes.jp/en/chief_editor/027.html
The Toyota page also has better illustration of different levels of drive automation.
Having them market Level 3 self driving 2 years after releasing that dangerous system makes me nervous.
Heh ...
It doesn't do lane changes or anything fancy, mind you, but I definitely like how it manages traffic jams; while normally I find them stressful, in adaptive cruise mode it's more like just idle time.
You do, of course, have to keep your hands on the wheel as that is the standard for anything built in in the US at the moment. If you got something like OpenPilot you could probably get away with hands-off.
The Honda E interior is a lot more interesting in my opinion. The Honda E may not have level 3 self-driving in traffic jam yet, but it does have a virtual aquarium.
Not sure this messaging is safe, much less the car itself.
No. That is the true benefit of Level3: you do not need to pay attention. The car is paying attention and will notify you when you must retake control.
Tesla's Level2 "autopilot" will not tell you before it drives into a wall. Tesla Drivers must constantly watch the road and ask themselves "Do I think my car is trying to kill me?". When they think "Yeah, my car appears to be trying to drive under that parked semi-truck. I don't we fit, I better take over".
Meanwhile Honda's level3 will say: "Please take over. I see a semi-truck and I don't think we'll fit."
> "There is a limit to the capabilities (e.g. recognition capability and control capability) of individual functions of Honda Sensing Elite," reads an official statement. "Please do not overestimate the capabilities of each Honda Sensing Elite function and drive safely while paying constant attention to your surroundings. Please remain in condition where you can respond to the handover request issued by the system, and immediately resume driving upon the handover request."
You cannot be watching a DVD and then immediately resume driving at request. It is not humanly possible in my opinion. There is a critical context-switching time that even the best of people will be slow to handle.
I'll admit I don't understand the L2/L3 difference - and maybe that is part of the messaging and the problem. Your demeaning quotes about Tesla drivers seem to fit 100% with how Honda wants you to behave in their car. They are telling you the system could fail and you need to be ready. But you can't "be ready" without already being ready - if you have to 'get ready' by quickly stopping your attention to the DVD and changing to the road, it may be too late.
L3: Your car will warn you and you'll have enough time to look up, realize what is happening, and say "oh shit"
Edit: it's a joke. LVL3 no way. https://www.youtube.com/watch?v=_hBwmFbpNCA
It's frustrating how technically illiterate the press is.
How many traffic intersections are there in the sky? Roundabouts? Wildlife? Domestic livestock?
You'd think that in a car, with its limited display size/resolution and glanceability required the UI/UX designers would spring for a more flat, colorful option so you can simply touch the orange square for music or something along those lines.
Not that touch controls in a car are a great UX in the first place, but you'd think they would have moved past the "everything is the same colored 3d rect with a glyph in the middle" phase of UI development.
So another batch of dangerous "half-self-driving" cars is hitting the streets, encouraging drivers to disengage from the wheel but ignoring the fact that distracted people tend to remain distracted much longer than their automated car desires.
There is no such thing as "Level 3 automation", because driving is a wholesome intelectual endeavor that cannot be compartmentalized. You cannot drop a distracted driver in a life or death trafic situation, an expect them to correct course efficiently.
Driving includes route planning, optical recognition of static and dynamic road conditions, trajectory prediction of other trafic participants, as well as being aware of the dynamic parameters of your own vehicle and predicting how your inputs will reflect into future trajectory. The brain is essentially a giant prediction machine, as explained yesterday by an article here on HN: https://news.ycombinator.com/item?id=26341218
This is a continuous loop where an attentive driver maintains state information from a large time period, from seconds to hours: this is slippery and curvy road section, a certain road sign was spotted a mile ago, the gray Honda following drives erratically, there is a kid with a ball and a cat on the sidewalk, the wheel is slightly tilted to right to match trajectory.
There is no way to partition this activity into "simple" self-driving tasks that can be automated and "complex" things a man can drop in and do. It's all or nothing. Once the man looks away from the road for half a minute, he has lost critical state information that affords him human level driving accuracy, and it would take minutes to warm up again, including gradual increase of control authority. His role in a "Level 3 self-driving car" is essentially that of the monkey to take the blame.
Another one of my ideas, do you know why Musk is building starlink ? To operate fleet of cars remotely all over the world by using remote operators and avoid paying NET connect to telco operators. With LEO orbit latency is piece of cake. Only thing is trsutworthy of that operators. The price for autdriving, will be payments for operators, calculated from local country HDP, to pay operators by local country "minimal" HDP wages. Problems with unemployment "solved".