2016 California Autonomous vehicle disengagement reports

[RubberToe]

Tesla and other automakers running self driving cars in California have all now submitted their 2016 testing results. With the most important metric being how often the driver needs to take control from the autonomous system.

Story here: Tesla Is Testing Self Driving Cars on California Roads

And the link to California test reports is here: https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/disengagement_report_2016

RT

 

[McRat]

GM started their testing with Nissan Leafs, then switched to Bolts in June. 2284 miles with 6 disengagements in November 2016.

 

[Green Pete]

Tesla and other automakers running self driving cars in California have all now submitted their 2016 testing results. With the most important metric being how often the driver needs to take control from the autonomous system.

Story here: Tesla Is Testing Self Driving Cars on California Roads

And the link to California test reports is here: https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/disengagement_report_2016

RT

Seems wet roads were an issue. Anyone know what follower and planner output means?

Edit: Looked at the dates again, wet roads were probably just the conditions around in the 8 day period that the cars were being used. Seems basically 100% of the disengagements were over the course of an 8 day period in October and the 20 miles in November (which we all have seen in the video) were without disengagement.

 

[NerdUno]

CA DMV Report Sheds New Light On Misleading Tesla Autonomous Drive Video - DailyKanban

 

[disagree]

CA DMV Report Sheds New Light On Misleading Tesla Autonomous Drive Video - DailyKanban

Good grief. Yikes.

 

[3Victoria]

Ugh -- I took a look at the linked 'article', under duress. I would say that Tesla did a remarkable job. The Planner/Follower errors seem like problems with communications between the course 'planning' software, which I assume interfaces with the mapping app, and the driving 'follower' software -- likely because they set it up for this demo. After they had debugged it in October, it looks like they had a clean run through the demo course without any disengagements. It is also pretty clear that they are testing this stuff elsewhere than California public roads, but set up the demo on public roads on purpose, and thus had to do a disengagements report. I am not seeing anything here distressing, other than more FUD from DK, which is a well-known anti-Tesla site and written by a well-known troll.

 

[disagree]

Ugh -- I took a look at the linked 'article', under duress. I would say that Tesla did a remarkable job. The Planner/Follower errors seem like problems with communications between the course 'planning' software, which I assume interfaces with the mapping app, and the driving 'follower' software -- likely because they set it up for this demo. After they had debugged it in October, it looks like they had a clean run through the demo course without any disengagements. It is also pretty clear that they are testing this stuff elsewhere than California public roads, but set up the demo on public roads on purpose, and thus had to do a disengagements report. I am not seeing anything here distressing, other than more FUD from DK, which is a well-known anti-Tesla site and written by a well-known troll.

You assert things for which you have no support and then conclude by poisoning the well.

Surely there is an internet term for the opposite of a troll...

Oh! Yes, it's FANBOY.

 

[ABC2D]

Ugh -- I took a look at the linked 'article', under duress. I would say that Tesla did a remarkable job. The Planner/Follower errors seem like problems with communications between the course 'planning' software, which I assume interfaces with the mapping app, and the driving 'follower' software -- likely because they set it up for this demo. After they had debugged it in October, it looks like they had a clean run through the demo course without any disengagements. It is also pretty clear that they are testing this stuff elsewhere than California public roads, but set up the demo on public roads on purpose, and thus had to do a disengagements report. I am not seeing anything here distressing, other than more FUD from DK, which is a well-known anti-Tesla site and written by a well-known troll.

Well, you might have outdone that "anti-Tesla site" with "well-known troll" by posting more FUD and BS in your one paragraph...

 

[McRat]

Look at the data submitted by Tesla from Tesla. Ignore any outside opinions/articles/conjectures. What do you see?

I see they are just starting testing in their home state. And that they did most of the testing over a very few days. The testing results were not leading the pack.

No spin necessary.

 

[RubberToe]

California will being allowing autonomous vehicles on the road later this year without human drivers...

California Says Autonomous Cars Don't Need Human Drivers

 

[Wolverinegeoff]

California will being allowing autonomous vehicles on the road later this year without human drivers...

California Says Autonomous Cars Don't Need Human Drivers

At least they won't be reporting any "human takeover" events anymore? Just robot doom vehicles taking their revenge for all the "Cats riding on Roombas" videos.

 

[McRat]

Let be honest. Have you seen California drivers? A car with no steering or pedals or even a computer control might do better than who's driving today. Just let it go and hope everybody gets out of the way isn't that much different than what we are doing now when we issue a CDL to somebody.

 

[Ulmo]

Let be honest. Have you seen California drivers? A car with no steering or pedals or even a computer control might do better than who's driving today. Just let it go and hope everybody gets out of the way isn't that much different than what we are doing now when we issue a CDL to somebody.

Are you comparing to the worst, average, or best drivers?

 

[McRat]

Are you comparing to the worst, average, or best drivers?

IMO, out of the 30+ states I've driven, California has the both the best and worst drivers. Our worst are pretty bad though. It has to do with our culture and geographic location.

But watching the decay occur due to cellphone use has been the most discouraging thing I have witnessed with our roads. It is costing us billions a year in lost time, higher fuel use, and more accidents.

 

[Wolverinegeoff]

IMO, out of the 30+ states I've driven, California has the both the best and worst drivers. Our worst are pretty bad though. It has to do with our culture and geographic location.

But watching the decay occur due to cellphone use has been the most discouraging thing I have witnessed with our roads. It is costing us billions a year in lost time, higher fuel use, and more accidents.

You should come to Florida. Between the young, "Fast and furious, oh look I got a Snapchat" crowd, and our older, retired, "it's always a blind spot when you're already blind so why check it?" crowd, we have an absolute hellscape on highways and surface streets. Plus....tourists!

 

[McRat]

You should come to Florida. Between the young, "Fast and furious, oh look I got a Snapchat" crowd, and our older, retired, "it's always a blind spot when you're already blind so why check it?" crowd, we have an absolute hellscape on highways and surface streets. Plus....tourists!

My first haul into Florida there was a flashing traffic warning display on the side of the 10 freeway right after Alabama, it said:

WELCOME TO FLORIDA
NOW SLOW DOWN

California has both temporary tourists and a lot of newcomers to the area, with high density traffic, lots of elevation changes, and bizarre freeway interchanges everywhere.

But I do need to give an Honorable Mention to both Florida and Las Vegas. Like you mentioned, both have a 50/50 mix of Fast n Furious and Slow n Senile drivers.

 

[Electroman]

I have no issues with Texas. Drivers here are great. !

 

[Wolverinegeoff]

My first haul into Florida there was a flashing traffic warning display on the side of the 10 freeway right after Alabama, it said:

WELCOME TO FLORIDA
NOW SLOW DOWN

California has both temporary tourists and a lot of newcomers to the area, with high density traffic, lots of elevation changes, and bizarre freeway interchanges everywhere.

But I do need to give an Honorable Mention to both Florida and Las Vegas. Like you mentioned, both have a 50/50 mix of Fast n Furious and Slow n Senile drivers.

Having lived in LA for a few years, I will agree that there is definitely some awfulness present. I wonder how Tesla Vision will handle crazy motorcyclists white-lining down the LA freeway at ridiculous speeds.

 

[gabl]

CA DMV Report Sheds New Light On Misleading Tesla Autonomous Drive Video - DailyKanban

The route map linked from reddit is incorrect. This is the correct map.
How could the witnesses be so wrong about the location?

 

[zmarty]

As an investor I feel that article is both accurate and very scary. A lot is riding on getting FSD working.

Tesla seems to be in the stone ages compared to Alphabet: Waymo Built a Secret World for Self-Driving Cars