Autonoma fordon

Här diskuterar vi elektriska fordon av alla slag

Moderator: Redaktion

Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Cruise och Waymo får tillstånd att gå live med Robotaxi (och kan därmed börja ta betalt).

https://docs.cpuc.ca.gov/PublishedDocs/ ... 694131.PDF
Användarvisningsbild
TiborBlomhall
Inlägg: 7297
Blev medlem: 04 mar 2014 22:03
Ort: Stockholm
Referralkod: ts.la/tibor72449

Re: Autonoma fordon

Inlägg av TiborBlomhall »

spacecoin skrev:
Cruise och Waymo får tillstånd att gå live med Robotaxi (och kan därmed börja ta betalt).

https://docs.cpuc.ca.gov/PublishedDocs/ ... 694131.PDF
...fast fortfarande måste mänsklig chaufför sitta bakom ratten beredd att ta över.

"The Drivered Deployment program is distinct in that it allows for fare collection and shared rides and requires a safety driver to be present in the vehicle."

https://www.cpuc.ca.gov/regulatory-serv ... -questions
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

TiborBlomhall skrev:
spacecoin skrev:
Cruise och Waymo får tillstånd att gå live med Robotaxi (och kan därmed börja ta betalt).

https://docs.cpuc.ca.gov/PublishedDocs/ ... 694131.PDF
...fast fortfarande måste mänsklig chaufför sitta bakom ratten beredd att ta över.

"The Drivered Deployment program is distinct in that it allows for fare collection and shared rides and requires a safety driver to be present in the vehicle."

https://www.cpuc.ca.gov/regulatory-serv ... -questions
Ja, men det dröjer nog inte länge förrän i alla fall Cruise är redo att köra helt utan om myndigheterna är det. Jag tippar på i år i SF för Cruise. Det är kul att det är ett litet race nu mellan Cruise och Waymo. Vi får se vem som kommer i mål först i riktig stadsmiljö.

Cruise har en liggande ansökan sedan i slutet på förra året:
https://www.cpuc.ca.gov/regulatory-serv ... ter-status
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Waymo har börjar köra interna tester i SF nu utan säkerhetsförare på dagtid:
https://blog.waymo.com/2022/03/taking-o ... y-bay.html
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Stoppa autonowashingen säger Electronic Engineering Times.
We hear repeated claims from AV advocates that using technology to replace humans as drivers will lead to safer roads. But where is the evidence? Moreover, where is the progress?
Privately owned passenger vehicles are not going to be autonomous, or self-driving, any time soon. As engineers, we know this, so why aren’t more of us speaking up and having this conversation?
The conclusion is obvious: There is no trend to autonomous operation in privately owned passenger vehicles, at least throughout this decade. Driver assistance, yes. Supervised automation, yes. But autonomous operation in consumer vehicles? No.
https://www.eetimes.com/we-really-need- ... confusion/
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

AutoX verkar inte så intresserade av säkerhet (dom heller):
https://archive.ph/C0JvF

Branschen är rätt rutten. Många oseriösa aktörer.
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

BMW 7 series (i7) får hands-free L2+ som Supercruise och L3 i kökörning som EQS senare:
https://www.thedrive.com/new-cars/45016 ... ving-fight
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »



VW kör om Tesla i förarstöd i EU?

Travel assist got huge improvements.
- automatic lanechange
- detection of roundabouts
- slowing prior to sign
- emergency lane in traffic jams
- sign recognition
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Första videon med Mobileye Drive (Level 4 robotaxi system). 40 min unedited - Jerusalem:


Målet är att vara live i Israel och Tyskland i år.
Användarvisningsbild
Zalman3
Teslaägare
Inlägg: 14855
Blev medlem: 29 mar 2016 18:28
Ort: Umeå

Re: Autonoma fordon

Inlägg av Zalman3 »

spacecoin skrev:
Uppdaterat första posten med en lista av spelare. Saknas någon?
John Deere Reveals Fully Autonomous Tractor at CES 2022
https://www.deere.com/en/news/all-news/ ... or-reveal/
Titanium S75D, svarta NG, svart alcantara, autopilot 2.0, air suspension
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Pony.ai becomes first self-driving firm to get taxi license in China
Pony.ai has received a license to operate 100 Robotaxis as traditional cabs in Nansha district, Guangzhou city, Guangdong province.

https://cnevpost.com/2022/04/24/nio-bac ... -in-china/

Pony.ai, a Chinese self-driving startup, is backed by Toyota Motor and NIO Capital.
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

50 autonoma gatsopare från WeRide:
https://autonews.gasgoo.com/m/70020239.html
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Self-driving lane on I-94 gets $130M from Ford and others:
https://www.mlive.com/public-interest/2 ... thers.html

Det här är ju så jävla dumt. Amerikaner vägrar verkligen åka kommunalt.
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Det är drygt två är sedan Stefan Seltz-Axmacher skrev sin välkända och ovanligt ärliga bloggpost i samband med att Starsky inte lyckades med sin Series-B-runda.

Fortfarande klockrena spaningar.

Förväntningar på AI:
hockey.png
"Five years later and AV professionals are no longer promising Artificial General Intelligence after the next code commit. Instead, the consensus has become that we’re at least 10 years away from self-driving cars.
It’s widely understood that the hardest part of building AI is how it deals with situations that happen uncommonly, i.e. edge cases. In fact, the better your model, the harder it is to find robust data sets of novel edge cases. Additionally, the better your model, the more accurate the data you need to improve it. Rather than seeing exponential improvements in the quality of AI performance (a la Moore’s Law), we’re instead seeing exponential increases in the cost to improve AI systems — supervised ML seems to follow an S-Curve."
sxurve.png
"The S-Curve here is why Comma.ai, with 5–15 engineers, sees performance not wholly different than Tesla’s 100+ person autonomy team. Or why at Starsky we were able to become one of three companies to do on-public road unmanned tests (with only 30 engineers)."
human.png
"If L1 is the line of human equivalence, then leading AV companies merely have to prove safety to be able to deploy. I don’t think I know anyone serious who believes this, but it is a possibility. If L2 is the case, the bigger teams are somewhere from $1–25b away from solving this problem. When big AV investors say that autonomy is an industry just for big companies, this is the bet that they’re making. If, however, L3 is the line of human equivalence it’s unlikely any of the current technology will make that jump. Whenever someone says autonomy is 10 years away that’s almost certainly what their thought is. There aren’t many startups that can survive 10 years without shipping, which means that almost no current autonomous team will ever ship AI decision makers if this is the case."

Finns även ett Autonocast-ansnitt som är en intervju om bloggposten: http://www.autonocast.com/blog/2020/3/2 ... f-autonomy

En annan intressant inlägg i debatten är detta från Gary Marcus (ex Uber ATG):
https://nautil.us/deep-learning-is-hitt ... all-14467/
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Dude, where's my self-driving car?
https://www.politico.com/newsletters/di ... r-00028703

"The biggest roadblock is not​ law. The federal government and several states have given the green light. The biggest roadblock is that robotics is hard."

"You could deploy autonomous cars today — if you limited where they went/what part of the city they could operate in, kept them under 25 mph, created their own lanes for them (think bus lanes), and only deployed them on good-weather days. The current sensor plus algorithm technology is capable of handling "normal" driving conditions, which is what your average car/driver experiences 99 percent of the time."
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Waymo utökar till downtown Phoenix:
"Calling all Downtown Phoenix residents! Our Waymo One Trusted Tester program is expanding. Help us shape the future of autonomous ride-hailing in your community by downloading the Waymo One app and expressing interest to ride with us there."

Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Who Will Win The Self-Driving Game?:
https://groundtruthautonomy.com/opinion ... ving-game/

"Risk: AV Edition is played simultaneously on multiple maps stacked like pancakes. Each of these maps represents different areas in which AV companies can operate different types of autonomous driving services, including but not limited to: sidewalk bots, robotaxi, last mile, middle mile, and trucking. (A future expansion pack could include privately owned autonomous cars, which is a very different game, with different mechanics.)"
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Motional börjar med autonoma IONIQ5 för att leverera mat:
https://motional.com/news/motional-and- ... nta-monica

(men givetvis ligger Tesla långt före alla andra) :)
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

ArgoAI live nu i två städer DRIVERLESS och DAGTID:
https://www.argo.ai/company-news/argo-a ... mi-austin/

Legacy auto (Ford, GM, VW, Mercedes) piskar skiten ur Tesla när det gäller autonomi. Mercedes har L3. De andra har L4.
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

"The misperception is apparent in Musk’s choice of words: “self-driving … that is better than human.” In developing an AV, engineers’ goal is to ensure that everything the vehicle does is what its human developers and operators want it to do. Unpredictable behavior by the vehicle must be extirpated. The goal in AV development is to ensure that humans so thoroughly determine how the vehicle performs that the vehicle itself can have zero autonomy.

AVs are in fact human-driven vehicles, though the operators are not in the vehicle, and they control it through programs rather than through a direct physical interface. These human operators must be biased – otherwise they will have no paying human customers. Except in niche applications or in services operated at a loss (such as Alphabet’s Waymo, and perhaps Tesla’s eventual robotaxis), they must not value safety so much that the vehicle never exceeds 20 mph. They must also value the customer’s experience more than they value the preferences of others outside the vehicle—if they don’t, they will lose their customers.

These human biases are business imperatives, and they come at a cost to safety that precludes the kind of safety gains that Musk and other AV promoters have predicted. AVs cannot become common until they can earn more revenue than they cost. And they can’t do that and prioritize safety over customer satisfaction."

https://www.marketwatch.com/story/elon- ... 1652877067
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Robot på Max... Verkar mer jippo än bra. "Yo! Sushi" hade typ det här i London för tio år sedan...
https://www.svd.se/a/Qy6jvQ/hamburgerre ... -med-robot

Är det någon som sett en?
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Yan LeCun säger det som jag har sagt i två år:
About the raging debate regarding the significance of recent progress in AI, it may be useful to (re)state a few obvious facts:
(0) there is no such thing as AGI. Reaching "Human Level AI" may be a useful goal, but even humans are specialized.
(1) the research community is making *some* progress towards HLAI
(2) scaling up helps. It's necessary but not sufficient, because....
(3) we are still missing some fundamental concepts
(4) some of those new concepts are possibly "around the corner" (e.g. generalized self-supervised learning)
(5) but we don't know how many such new concepts are needed. We just see the most obvious ones.
(6) hence, we can't predict how long it's going to take to reach HLAI.
I really don't think it's just a matter of scaling things up.
We still don't have a learning paradigm that allows machines to learn how the world works, like human and many non-human babies do.
Some may believe that scaling up a giant transformer trained on sequences of tokenized inputs is enough.
Others believe "reward is enough".
A few others believe that explicit symbol manipulation is necessary.
A few don't believe gradient-based learning is part of the solution.
I believe we need to find new concepts that would allow machines to:
- learn how the world works by observing, like babies.
- learn to predict how one can influence the world through taking actions.
- learn hierarchical representations that allows long-term predictions in abstract representation spaces.
- properly deal with the fact that the world is not completely predictable.
- enable agents to predict the effects of sequences of actions so as to be able to reason and plan
- enable machines to plan hierarchically, decomposing a complex task into subtasks.
- all of this in ways that are compatible with gradient-based learning.
The solution is not just around the corner.
We have a number of obstacles to clear, and we don't know how.
Användarvisningsbild
RagWal
Teslaägare
Inlägg: 1214
Blev medlem: 06 dec 2018 19:29
Referralkod: ts.la/ragnar71607

Re: Autonoma fordon

Inlägg av RagWal »

spacecoin skrev:
Robot på Max... Verkar mer jippo än bra. "Yo! Sushi" hade typ det här i London för tio år sedan...
https://www.svd.se/a/Qy6jvQ/hamburgerre ... -med-robot

Är det någon som sett en?
Vet inte om det var exakt samma, men motsvarande finns på Waipo i centrala Sthlm.
Kockarna lastade på tallrikar i köket och den lallade sen iväg till rätt bord där personal serverade från den. Klarade att ta sig fram bra i gångfart och stannade / saktade in när folk sprang förbi eller stod i vägen.
Oklart hur mycket som sparades i tid etc för personalen men jag kan tänka mig att det iaf blir mindre spring till köket.
Såklart lite uppmärksamhet också.
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

Hittade denna utmärkta förklaring på varför vision-only-system för autonomi (speciellt utan hd-kartor och geofence) sannolikt är decennier bort:
In general. ML systems, like, for example, the one Tesla is using to make sense of flat grid pixels to decide what is an obstacle and where it is in 3D space - are statistical blackbox. You feed a bunch of examples with metadata, some "magic" happens inside, and if everything goes well, the system will give you a right answer with good probability, say 98% of the time. I guess I don't need to explain why trusting your life to a system that 1 time out of 50 could fail to recognise a solid obstacle on your path and drive you through - isn't such a great idea.

But the interesting part is that even a very reliable ML box can't be truly validated to be working reliably, sometimes they fail spectacularly in seemingly simple cases and we don't know why. In general humans have very little understanding of what is happening inside ML systems - they are just millions of interconnected weights which are very difficult to analyse or rationalise why the net behaves in a certain way.

While ML techniques improve and systems get better and more reliable, for it to be a safety-critical component some major theoretical breakthrough is needed which will bring a robust validation and analysis of ML system. Good chance it will have to be something entirely different from the current ConvNet approach and nothing like this seems to be brewing in academia or industry as far as I know.

While this hasn't happened, autonomy companies are forced to rely on direct measurement sensors like LIDARs which don't require ML to discern drivable surfaces and obstacles. They still use ML of course, but not in the safety-critical path - for example in classification and prediction subsystem. This way they can still break or drive around an obstacle even if they don't understand what they are seeing, while vision-only system will happily drive the vehicle into such obstacle as it wasn't there.
Användarvisningsbild
spacecoin
Teslaägare
Inlägg: 5880
Blev medlem: 30 aug 2019 19:30

Re: Autonoma fordon

Inlägg av spacecoin »

New regulations for self-driving cars pass German Upper House (Bundesrat)


Bra nyheter för VW/Argo och MobilEye.
Skriv svar