Navigant: GM Cruise, Waymo Lead Tesla And Others On Full Autonomy

[TMC Staff]

Researchers believe Tesla is underperforming in autonomous vehicle development. Each year, Navigant Research analyzes the major players in the Autonomous Vehicle space. The research firm ranks each company on their program strategy and execution in bringing fully autonomous vehicles to market in some fashion. According to Navigant, vendors are rated on 10 criteria: vision, go-to...
[WPURI="https://teslamotorsclub.com/blog/2019/03/20/tesla-autonomous-lags-gm-ford-waymo/"]READ FULL ARTICLE[/WPURI]

 

[wjhepworth]

Navigant is silly. How many of these vehicles are actually being sold to customers? Correct me if I am wrong but it seems they are all operating in controlled fleets. Far cry from the real world.

 

[xtermz]

I don’t know about GM, but I do consistently see Waymo here in the Bay Area. They’re usually autonomous minivans driving on local roads, respecting traffic lights and signs. They drive pretty aggressively, but they also seem to brake for even the slightest adjustment, instead of making use of the padding between the car in front. Kind of annoying if you’re behind a Waymo.

 

[Jtrader]

You can’t compare those that use Lidar, because that isn’t a realistic solution.

 

[vickh]

why isn't Lidar a realistic solution. Isn't AP1 similar?

Waymo is one reason I'm considering Pacifica PHEV to replace our Highlander SUV. The ACC is only available on the limited version, thats over $40K. I'm hoping they open up the tech to customers one day, rather then fleets

 

[Dutchie]

Strange, this technology is hardly out there. Besides, how can they know how far ahead each manufacturer is? A lot of these technological developments are well guarded secrets.

 

[cwerdna]

Strange, this technology is hardly out there. Besides, how can they know how far ahead each manufacturer is? A lot of these technological developments are well guarded secrets.

Although we had no idea for the Navigant's previous year's report for Apple, we certainly have some idea and for many of the others and it's been discussed over and over in Autonomous Car Progress.

Look at Autonomous Car Progress what Google showed in 2012. Look at what Nissan showed: allowing a journalist along in 2017.

Have you looked at the California public road disengagement reports at Testing of Autonomous Vehicles for Tesla vs. Waymo/Google and Cruise Automation for each year? Keep in mind that Tesla HQ is in Silicon Valley and there's a huge body of software engineering talent here. UPDATE: Disengagement Reports 2018 – Final Results has a summary about the most recent results.

(Hint: You will find Tesla's results comical vs. Waymo.)

Tesla has missed Elon's promised dates over and over (e.g. cross country trip that was supposed to have been completed years ago).

I guess you're unaware of Waymo has launched its commercial self-driving service in Phoenix — and it's called 'Waymo One' and Why testing self-driving cars in SF is challenging but necessary. The latter took reporters along for My Herky-Jerky Ride in General Motors' Ultra-Cautious Self Driving Car.

If Tesla was even as far as say Nissan (not a leader) and why is it that they keep shipping out software that can't reliably stop for stopped vehicles (e.g. firetrucks and police cars) and can't read stop lights nor reliably detect pedestrians and bicyclists? Since Tesla's about hype, how come they've yet to demonstrate that they're at least equivalent to Nissan or Cruise Automation w/reporters on board? Or, how about a closed beta taxi program like Waymo had?

 

[Rogstar]

I don’t know about GM, but I do consistently see Waymo here in the Bay Area. They’re usually autonomous minivans driving on local roads, respecting traffic lights and signs. They drive pretty aggressively, but they also seem to brake for even the slightest adjustment, instead of making use of the padding between the car in front. Kind of annoying if you’re behind a Waymo.

Do they drive outside of "mapped" urban areas?
They don't seem to account for the upcoming upgrade from Tesla.

 

[Rogstar]

Fleets operating in cities are in a controllable, limited environment.
Can the waymo and other systems be given a totally random destination outside of their normal operating area and successfully navigate to that point in any weather conditions?
I think that's what we're going to see from the Tesla hardware update in a few months.
Hopefully there will be a reassessment of the situation at that point?

 

[Rogstar]

Although we had no idea for the Navigant's previous year's report for Apple, we certainly have some idea and for many of the others and it's been discussed over and over in Autonomous Car Progress.

Look at Autonomous Car Progress what Google showed in 2012. Look at what Nissan showed: allowing a journalist along in 2017.

Have you looked at the California public road disengagement reports at Testing of Autonomous Vehicles for Tesla vs. Waymo/Google and Cruise Automation for each year? Keep in mind that Tesla HQ is in Silicon Valley and there's a huge body of software engineering talent here. UPDATE: Disengagement Reports 2018 – Final Results has a summary about the most recent results.

(Hint: You will find Tesla's results comical vs. Waymo.)

Tesla has missed Elon's promised dates over and over (e.g. cross country trip that was supposed to have been completed years ago).

I guess you're unaware of Waymo has launched its commercial self-driving service in Phoenix — and it's called 'Waymo One' and Why testing self-driving cars in SF is challenging but necessary. The latter took reporters along for My Herky-Jerky Ride in General Motors' Ultra-Cautious Self Driving Car.

If Tesla was even as far as say Nissan (not a leader) and why is it that they keep shipping out software that can't reliably stop for stopped vehicles (e.g. firetrucks and police cars) and can't read stop lights nor reliably detect pedestrians and bicyclists? Since Tesla's about hype, how come they've yet to demonstrate that they're at least equivalent to Nissan or Cruise Automation w/reporters on board? Or, how about a closed beta taxi program like Waymo had?

You don't seem to be paying much attention to developments?
You quote supposed deficiencies in the Tesla system based on year old incidents and data, but conveniently forget both the waymo "pedestrian incident" and the imminent Tesla hardware update.
May I suggest you listen (again?) to the Elon Musk ARK interview? Or are you just ignoring the contents?

 

[electronblue]

waymo "pedestrian incident" and the imminent Tesla hardware update.

Except that was Uber.

May I suggest you listen (again?) to the Elon Musk ARK interview? Or are you just ignoring the contents?

I recommend nobody takes cues from Elon Musk on FSD. The history just isn’t there.

 

[Rogstar]

Except that was Uber.



I recommend nobody takes cues from Elon Musk on FSD. The history just isn’t there.

Has he, or anyone from Tesla ever described autopilot in its present form as FSD?

Have you listened to the ARK interview?

 

[OPRCE]

Interesting to trace how Tesla has fared on the Navigant Leaderboard over the last few years:

March 2017: Tesla was still considered an AV contender, with a decent strategy and execution, obviously based mainly on the performance of AP1 and the faked demo video of Nov.2016:



March 2018: Competition heats up while the hot mess of EAP throughout 2017 is judged ... consequently the arse drops out of Tesla's rating as the industry recognises Musk's fantasies on execution largely dropped out of his arse:

March 2019: With strategy now also moving in reverse Tesla is relegated to being the tail-end Charlie of AV, challenging only to customers who paid for FSD years before yet with increasing frustration await its appearance and other companies like Apple with no tangible product:

By contrast, MobilEye, which Musk improbably claimed in 2016 was holding him back from releasing some amazing cross-country FSD, has now deservedly advanced to #5 in the overall ranking, by demonstrating themselves as a serious AV company who know WTH they are talking about.

Predictions:

1. At this rate of "progress", and given the recent redefinition of FSD to [at least legally] mean a fraction of what it previously did, by March 2020 Tesla will be alone in the 'followers' category, if it even makes it onto this board at all, seeing how FSD is promised to remain L2 with incessant nags (and an apparently insatiable affinity for parked firetrucks) for the foreseeable future.

2. HW3 & FSD will fail to pull off the rapid series of miracles needed inside the next 12 months to reverse this trend. Instead it will advance with all the grace and speed of an arthritic frog, as EAP development has done to date. If it achieves L3 approval by Jan.2023 without a sensor upgrade, that will be amazing.

3. The vaunted Tesla Network of driverless RoboTaxis [all mention of which was recently and silently erased from their website] will not be seen this side of 2028, and definitely not ever on the current HW2.5 sensor suite.

 

[Rogstar]

"Prediction"?
A tweet announcing completion of a coast to coast fully autonomus trip within a few months of HW3 Initiation, either by Tesla, or an owner.

Added to the long list of "Tesla can't / Tesla won't" FUD.

 

[cwerdna]

You quote supposed deficiencies in the Tesla system based on year old incidents and data,

So, you're saying that Teslas now can reliably detect stopped vehicles and not collide into them and do the same for pedestrians and bicyclists? How about reading and acting upon traffic lights?

Maybe you didn't bother looking at the CA autonomous vehicle disengagement reports. I'll point you to them directly.

Tesla:
2015: https://www.dmv.ca.gov/portal/wcm/c...la_disengagement_report.pdf?MOD=AJPERES&CVID=
2016: https://www.dmv.ca.gov/portal/wcm/c...a_disengage_report_2016.pdf?MOD=AJPERES&CVID=
2017: https://www.dmv.ca.gov/portal/wcm/c...9109-0c187adebbf2/Tesla.pdf?MOD=AJPERES&CVID=
2018: https://www.dmv.ca.gov/portal/wcm/c...596f2625a7f/TeslaMotors.pdf?MOD=AJPERES&CVID=

Compare these to Google/Waymo:
https://www.dmv.ca.gov/portal/wcm/c...le_disengagement_report.pdf?MOD=AJPERES&CVID=
https://www.dmv.ca.gov/portal/wcm/c...o_disengage_report_2016.pdf?MOD=AJPERES&CVID=
https://www.dmv.ca.gov/portal/wcm/c...-97f6f24b23cc/Waymofull.pdf?MOD=AJPERES&CVID=
https://www.dmv.ca.gov/portal/wcm/c...a5b8-eeca5477ace5/Waymo.pdf?MOD=AJPERES&CVID= - unfortunately 2018 doesn't have a nice summary, so you'll want to look at UPDATE: Disengagement Reports 2018 – Final Results.

but conveniently forget both the waymo "pedestrian incident"

You mean Uber's?

May I suggest you listen (again?) to the Elon Musk ARK interview? Or are you just ignoring the contents?

Not worth wasting my time on it. Elon's been so far off the mark w/his predictions so many times on so many things.

If Tesla's on par w/some of the leaders or non-leaders, given their propensity towards hype, they should be able to have journalists along to show their progress. They should actually be doing autonomous vehicle testing in California, their home state w/such a large body of software engineering talent and be able to demonstrate a high number of miles covered per year in a mixed environment with a competitive disengagement rate.

If they do that and are able to handle complex city environments like Cruise Automation and run a robotaxi service, even if limited to a couple hundred folks under NDA like Waymo had earlier, that'd be more evidence they're not far behind or behind at all.

 

[cwerdna]

"Prediction"?
A tweet announcing completion of a coast to coast fully autonomus trip within a few months of HW3 Initiation, either by Tesla, or an owner.

This was supposed to happen by end of 2017. It still hasn't.

Tesla wants new self-driving tech to autonomously road trip from LA to New York
Tesla still hasn’t completed its coast-to-coast autonomous drive
Tesla’s autonomous ‘coast-to-coast’ drive on hold while Autopilot team focuses on safety

 

[OPRCE]

"Prediction"?
A tweet announcing completion of a coast to coast fully autonomus trip within a few months of HW3 Initiation, either by Tesla, or an owner.

Added to the long list of "Tesla can't / Tesla won't" FUD.

Coast-to-coast sporting a large orange stuck in the steering-wheel to satiate the nags, eh?

Musk has stated FSD will remain L2 until approved by regulators after billions of fleet miles for validation, which even in his optimistic imagination was not projected to be before end of 2020.

Realism =/= FUD

 

[electronblue]

Has he, or anyone from Tesla ever described autopilot in its present form as FSD?

Have you listened to the ARK interview?

I believe I have listened to pretty much every Elon Musk talk on autonomoy since 2015 or so. I don’t believe him anymore.

That doesn’t mean Tesla can’t make progress on AP2 and beyond, of course they will. But Musk isn’t a reliable source on this one, says history.

 

[OPRCE]

Has he, or anyone from Tesla ever described autopilot in its present form as FSD?

Have you listened to the ARK interview?

Quoting Musk from that podcast you mention:
“Sure. We have - we already have full self-driving capability on highway. So from highway on ramp to highway exit, including passing cars and going from one highway interchange to another, full self-driving capability is there.”
... which sounds suspiciously like NoAP without confirmations, so I would say, yes, he continues to recklessly conflate AP with FSD, then effortlessly pivots to instantly pin the blame for the resultant deaths on those who were naïve enough to believe him.

Correction, the quote is from the Q4.2018 earnings call, transcript here.

 

[Rogstar]

Quoting Musk from that podcast you mention:
“Sure. We have - we already have full self-driving capability on highway. So from highway on ramp to highway exit, including passing cars and going from one highway interchange to another, full self-driving capability is there.”
... which sounds suspiciously like NoAP without confirmations, so I would say, yes, he continues to recklessly conflate AP with FSD, then effortlessly pivots to instantly pin the blame for the resultant deaths on those who were naïve enough to believe him.

"CAPABILITY"!
Do you think that HE hasn't used HW3?

"Resultant deaths"?
That's bs and you know it.