[don shaw]

An article at 'Seeking Alpha' reviews and cites formal data submitted to California by companies testing self driving cars in CA.

Data shows tesla =1 disengagement (human intervention) every 3 miles, GM/Cruise = 1 disengagement every 54 miles average for year, but reaching nearly 1 in 400 in last month of testing with Bolts.. Google leads

here's the link:
http://seekingalpha.com/article/404...ams?li_source=LI&li_medium=liftigniter-widget

The title is 'The Tesla Self-Driving Dreams Are Just That... Dreams'
No wonder there are 1006 comments!

Don

 

[Sly Bob]

I just wish I could buy one...

 

[DonC]

The recent advances by GM are impressive.

 

[accelerus]

I find it funny they seem to completely ignore the fact that Tesla has over 1.3 Billion autopilot miles under it's belt (as of Nov 2016). As of the middle of 2015 Waymo only has about 1 million miles under it's belt (last data point I could find)

That's like saying a boxer retired undefeated but only was ever in 1 fight in his life. And any form of semi-autonomous driving from Waymo/Bolt are all just in the testing phase, while Tesla has products you can buy today. It's not perfect by any means, but it's what we have right now. All three are hard at work of course working on constant improvement, but after reading that the vibe I got was the author is secretly pissed he can't go out and get a Tesla.

The "plan" is that Autonomous Bolts aren't due to hit the road till 2018 sometime. Will Tesla just sit on it's hands for 2 years and do nothing in that time?

The author, a supposed research analyst, apparently didn't bother to do that much research before badmouthing Tesla. For all his complaints it still doesn't change the fact if you want autonomous anything, Tesla is the only game in town for now. You can talk GM/Google up all day long, but until you can go buy it, it doesn't mean anything. This is just like those articles you see every year or two about "an extraordinary new battery tech that will revolutionize the industry, it charges in 5 minutes, holds 100Kwh, and only weighs 100lbs... doesn't matter how awesome it was in the lab, until it's something consumers can buy it doesn't mean anything.

 

[AndY1]

I find it funny they seem to completely ignore the fact that Tesla has over 1.3 Billion autopilot miles under it's belt (as of Nov 2016). As of the middle of 2015 Waymo only has about 1 million miles under it's belt (last data point I could find)

You can't compare Tesla's highway Lane Keep + ACC assistance with Google's autonomous driving.

 

[Bluesideup]

You really gotta love news reports like this that are so bad they are good.

First Tesla only submitted their self driving cars data for the new generation cameras and radars that they had just installed on the Model X that was being tested back in October/November when this report was sent to California. Second this only counts the driving that was done in California. Third this was on a production car during its testing phase with software that was still under construction. Do you want to bet how far Tesla's software has come in the last couple months or where it will be in the next couple?
The fact that Tesla actually has cars out in the wild with the capabilities to drive autonomously once it is enabled all the while mapping and learning during its hundreds of millions of miles of road use and you really think that a car no one can buy using equipment that isn't even put into the vehicle is somehow ahead of Tesla?

 

[Bluesideup]

You can't compare Tesla's highway Lane Keep + ACC assistance with Google's autonomous driving.

Sure you can. The only thing you can't compare is Google's autonomous driving because the only people that can use Google's autonomous driving is Google.

Here is a video of what the current production car is capable of with the new autopilot sensors once the software has been activated. Mind you this is back in November of last year and I am sure they have done A LOT more testing since then:

 

[don shaw]

Tesla stopped using it's initial supplier of autonomous components (or complete system?) and went to a new system. Did that negate the experience. performance results and data obtained with the millions of miles logged with the old system? Similarly, Cadillac delayed their (GM) system.

i would love to have an autonomous vehicle, especially for my wife who has a hard time driving at night, and I am happy to see the progress on everyone's part. But it is hard for me to understand what is really going on after the change in system(s).

 

[don shaw]

digging further, i went to the autonomous vehicle thread on the tesla motor club forum, and found the following post. There are many more pro and con comments there re the state of Tesla self driving - sorry i cannot seem to add a link, but recommend you go there.

Here's the post:

NerdUno, Jan 30, 2017
I just sent the following note to Tesla, but wanted to alert others to some of the problems we have experienced with AP2...

We have 17.3.2 and have used AutoSteer extensively on a number of interstate highways since its initial release. Primitive would be an understatement. Here's what we have encountered. On a straight road with no cars, AP2 appears to work satisfactorily at 45 MPH. On a road which curves leaving the vehicle driving into the sun even with human-readable lane markings that remain easy to decipher with the naked eye, the car is worse than dangerous. The car immediately zig zags from lane to lane with no realization that vehicles may be in the other lanes, even beside the car. On highways with exit ramps, AP2 always follows the exit ramp if you are in the right lane even without a right turn signal and even with clearly marked dotted lines on left side of the lane and even if following a vehicle proceeding straight ahead. AP2 still slams on the brakes when approaching almost any (stationary) overhead sign on an interstate.

I appreciate that AP1 code needed tweaking for the new hardware, but AP2 really leaves the impression that the developers started over... and have not progressed even to the level of what most would consider safe driving. There really should be more warnings and alerts regarding documented problems for those that are expecting a safer driving experience. A traffic fatality would be a huge setback for Tesla, not to mention the poor driver.
Informative x 8 Helpful x 2 Like x 2 Love x 1

 

[Bacardi]

Tesla = Production, very few could look at brand new AP2.0-equiped tesla and an non-AP older Tesla and point out the AP tech...
Bolt EV = Not production, huge roof arrays of equipment and unknown costs. Even if left the equipment arrays and just put a fancy trim cover it and charged $100K there would be people who would buy it, but GM may not be releasing any autonomous vehicles for private ownership over the Gen1 Bolt EVs life cycle; this is in part due to regulation vs GM's ability...But I do expect GM to sell these to fleets...

 

[don shaw]

GM (Mary Barra) said they were to produce (approximately) 50 autonomous Bolts on the assembly line in early 2016 for testing in Michigan (including
severe winter weather). Has anyone seen theses? I wonder if the roof arrays will look similar to those on the Cruise text vehicles, or if they will be more refined?

2011 Red Volt # 574 - lease expired
2014 Ashen Grey Volt - lease expired
2017 Siren Red
Never any maintenance on any except free oil change - but Red volts are faster

 

[theo1000]

Personally I can't wait for these autonomous vehicles to show up. I'm essentially blind in low light conditions and anything outside the well lit areas is a blur. I have to avoid night, esp. evening driving due to this. There are many like me with various complications who really are not optimally safe on the roads in all conditions.

That said....

I purchased a top of the line BMW I3 REX last year for my wife just to test the waters. And I have to say even things like the ACC are extraordinarily primitive and finicky. In the more robust weather conditions we have put it through it can not stand up to even little things like a little fog or dust from lawn mowing. A humidity change from 30% to 60%, something that can happen when you drive a couple of miles around easily disables it. The autonomous parallel parking is fairly primitive as well.

No doubt in production cars, TESLA is way ahead of BMW and pushing other manufacturers as a whole to advance the tech. But reading the reports I'm not sure if all this is anything more than Generation 0.1 type technology, including TESLA.

My own opinion is
2-3 disengagements per trip is Gen 0.1 tech or even less . Not sure we are even at this place. If your copier did this you would not have bought it even 50 years ago.

The end should be 2-3 disengagements per year in all weather conditions. We are verily at the bottom of the mountain and arguing over molehills.

 

[Bluesideup]

digging further, i went to the autonomous vehicle thread on the tesla motor club forum, and found the following post. There are many more pro and con comments there re the state of Tesla self driving - sorry i cannot seem to add a link, but recommend you go there.

Here's the post:

NerdUno, Jan 30, 2017
I just sent the following note to Tesla, but wanted to alert others to some of the problems we have experienced with AP2...

We have 17.3.2 and have used AutoSteer extensively on a number of interstate highways since its initial release. Primitive would be an understatement. Here's what we have encountered. On a straight road with no cars, AP2 appears to work satisfactorily at 45 MPH. On a road which curves leaving the vehicle driving into the sun even with human-readable lane markings that remain easy to decipher with the naked eye, the car is worse than dangerous. The car immediately zig zags from lane to lane with no realization that vehicles may be in the other lanes, even beside the car. On highways with exit ramps, AP2 always follows the exit ramp if you are in the right lane even without a right turn signal and even with clearly marked dotted lines on left side of the lane and even if following a vehicle proceeding straight ahead. AP2 still slams on the brakes when approaching almost any (stationary) overhead sign on an interstate.

I appreciate that AP1 code needed tweaking for the new hardware, but AP2 really leaves the impression that the developers started over... and have not progressed even to the level of what most would consider safe driving. There really should be more warnings and alerts regarding documented problems for those that are expecting a safer driving experience. A traffic fatality would be a huge setback for Tesla, not to mention the poor driver.
Informative x 8 Helpful x 2 Like x 2 Love x 1

Digging further I searched for sites that let us know what the consumers for the autonomous Bolt and Google cars were capable of. Funny I found zero references but maybe that is because there isn't a single working production car that either has to offer. We don't even know if they can even stay in a lane or exit a ramp because there is no car out there for anyone to test and I guarantee that none of the engineers are going to talk negative about what the cars are can't do. So, in all reality the Tesla is leaps and bounds ahead of a product that we don't even truly know can drive by itself.

Tesla is taking their autopilot slow and unlike all the others they are learning at an exponential rate as every car they sell provides information and feedback on the autopilot system. With the advent of the model 3 and the hundreds of thousands of drivers that will hit the road. Tesla will accumulate the hundreds of Million to Billions of miles quickly taking them far ahead of any competitor that has sensors installed on the top of their cars driven by professional drivers.

I can attest that the AP in the older tesla's works very well and with the advent of more cameras better sensors and 100X faster computer they put into the newer ones it will work and function even better.