Musk endeavours

[odysseus2000]

redsturgeon
http://uk.businessinsider.com/tesla-elo ... ?r=US&IR=T

Apparently and somewhat unsurprisingly, it seems like some shortcuts were taken to (almost) achieve the promised numbers.

John

Excellent piece of how journalist slant stories, creating an impression of shoddy testing & corner cutting using terms like "apparently", "insider", "industry expert" tailored to please Tesla bears

A journalist who has no personal knowledge or understanding of the process of manufacturing, relies on the comments of one worker with unknown motivation, and then finds someone who she says is an industry expert to say that what the worker has told her, if it is true, is not something done in other manufacturing plants.

She then calls a Tesla spokesperson who say all cars pass safety tests but doesn't then get into specifics.

Folk will believe this stuff if they want to & one can expect more similar stuff, but until one has safety data from reliable sources like official testing agencies or a large number of owners complaining it is just slanted opinion.

One should not forget that Tesla cars have received the highest safety numbers in all the crash tests performed by the US testing agencies.

Regards,

[Meatyfool]

http://uk.businessinsider.com/tesla-elon-musk-orders-engineers-to-stop-brake-and-roll-testing-2018-6?r=US&IR=T

Apparently and somewhat unsurprisingly, it seems like some shortcuts were taken to (almost) achieve the promised numbers.

John

One can never say never, but as Ody points out, this could easily be something other than it appears.

A disgruntled supervisor who has the authority to switch the "critical" on and off for example? Complete the test, swap it to off, take photo, swap it back to on, test the next car.

Wouldn't be surprised if a brown envelope swapped hands either.

Could have happened, but I cannot fathom that Tesla would permit reputation and safety risk to suffer just for a few hours of production.

Meatyfool..

[PeterGray]

Folk will believe this stuff if they want to

They will, just as others will come up with explanations as to why it couldn't possibly be the case.

I agree the story needs reading with a pinch of salt, but accepting it without question is no more likely to lead to false understanding than rejecting it out of hand.

Peter

[BobbyD]

A disgruntled supervisor who has the authority to switch the "critical" on and off for example?

If this is the case, again, it's a wonder Tesla ever makes any cars given that half of its workforce appears to consist of saboteurs bent on bringing the company to its knees.

Could have happened, but I cannot fathom that Tesla would permit reputation and safety risk to suffer just for a few hours of production.

Sounds entirely consistent with a company which sells an overdeveloped cruise control as 'Full self-driving hardware'.

[BobbyD]

Could have happened, but I cannot fathom that Tesla would permit reputation and safety risk to suffer just for a few hours of production.

Sounds entirely consistent with a company which sells an overdeveloped cruise control as 'Full self-driving hardware'.

Speaking of which...

Tesla agrees to settle class action over Autopilot billed as 'safer'

NEW YORK (Reuters) - Tesla Inc (TSLA.O) on Thursday reached an agreement to settle a class action lawsuit with buyers of its Model S and Model X cars who alleged that the company’s assisted-driving Autopilot system was “essentially unusable and demonstrably dangerous.”

The lawsuit said Tesla misrepresented on its website that the cars came with capabilities designed to make highway driving “safer.”

..

Under the proposed agreement, class members, who paid to get the Autopilot upgrade between 2016 and 2017, will receive between $20 and $280 in compensation. Tesla has agreed to place more than $5 million into a settlement fund, which will also cover attorney fees.

- https://uk.reuters.com/article/uk-usa-t ... KKBN1JU0MD

Tesla's Autopilot keeps crashing into parked cars. Here's why.

Another week, another Tesla crashing into a stationary vehicle that just came out of nowhere.

But that's the thing. The electric car's semi-autonomous driving assistance feature, known as Autopilot, has this very situation — parked cars seemingly coming out of nowhere — written into the manual. It's a known limitation of the driver assistance tool.

...

Here's what the manual says about this driving situation:

"Traffic-Aware Cruise Control cannot detect all objects and may not brake/decelerate for stationary vehicles, especially in situations when you are driving over 50 mph (80 km/h) and a vehicle you are following moves out of your driving path and a stationary vehicle or object is in front of you instead. Always pay attention to the road ahead and stay prepared to take immediate corrective action. Depending on Traffic-Aware Cruise Control to avoid a collision can result in serious injury or death."

...

The issue at play here isn't so much that the car is really bad at tracking parked cars and stationary objects out of view (that's a separate shortfall of Autopilot to get into), but that Tesla isn't making it clear to its drivers that the car can't handle or anticipate this fairly typical driving scenario.

In an email statement, a Tesla spokesperson defended using Autopilot appropriately and reminded drivers that the onus is always on them, even when using the semi-automated tool.
"When using Autopilot, drivers are continuously reminded of their responsibility to keep their hands on the wheel and maintain control of the vehicle at all times. Tesla has always been clear that Autopilot doesn’t make the car impervious to all accidents, and before a driver can use Autopilot, they must accept a dialogue box which states that ‘Autopilot is designed for use on highways that have a center divider and clear lane markings,’" the statement read.

- https://mashable.com/2018/05/30/tesla-a ... urope=true

If only they really did have full self-driving hardware https://www.youtube.com/watch?v=scWkuh4I8HI

[BobbyD]

oops, wrong video: https://www.youtube.com/watch?v=MpQ73bSFcyk

[odysseus2000]

Rebuttal of the Business Insider lack of testing story.

(FUD= Fear, Uncertainty & Doubt, usual salesperson tactic to convince a punter to buy what the salesperson is selling not some better & lower cost alternative)

https://cleantechnica.com/2018/07/04/bu ... s-pravduh/

[BobbyD]

Rebuttal of the Business Insider lack of testing story.

(FUD= Fear, Uncertainty & Doubt, usual salesperson tactic to convince a punter to buy what the salesperson is selling not some better & lower cost alternative)

https://cleantechnica.com/2018/07/04/bu ... s-pravduh/

It's hard to argue with such a dipassionate and objective piece. Mainly because it's really grating and hard to read.

I'm also not sure that

But Tesla has never been one to conform just for the sake of conformance. It has never built its products, software, experiences, factories, or tests according to a standard just because that’s how it has always been done.

is a great line of defence for a company withe Tesla's lets say controversial safety record.

[odysseus2000]

BobbyD
is a great line of defence for a company withe Tesla's lets say controversial safety record.

Since Tesla launched its first car, how many reports have you seen of the deaths in cars made by Ford, GM, GE, Toyota, Volvo ...

Of the ten people who will on average die on UK roads today, how many will involve a Tesla and how many will be report. Hint, if a Tesla isn't involved there will be no report, if a Tesla is it will make the main media channels

All of the crash and injury information is freely available as are the clear and document cases when Toyota cars killed people due to defective throttles that stuck open, but one does not see reports on them. E.g. what is the chance of being killed if the car(s) you own are involved in a crash. That would be interesting, but who bothers to research it.

Meanwhile how many people have suffered medical breathing conditions from breathing diesel and petrol exhaust in the same time period? Many health professionals cite internal car engine exhaust as a major cause of breathing troubles.

Sure Tesla has to be accountable, but one should not forget that we don't hold other car makers to anything like the same standard.

Regards,

[tjh290633]

Do we not? What is the incidence of problems with Tesla models as a percentage of vehicles in use?

I can foresee a mass recall because of failings with the automatic pilot software. Several have been reported already. Ultimately there could be prohibition.

TJH

[odysseus2000]

Do we not? What is the incidence of problems with Tesla models as a percentage of vehicles in use?

I can foresee a mass recall because of failings with the automatic pilot software. Several have been reported already. Ultimately there could be prohibition.

TJH

According to Wikipedia there have been 4 autonomous vehicle deaths:

https://en.wikipedia.org/wiki/List_of_a ... fatalities

3 of these have been Tesla.

Tesla have sold something less that 200,000 cars in the US as I understand it, this being the threshold for the loss of the incentive which is still in force in the US as far as I know.

One presumably wants to know the death rate per mile or similar statistic across various marques to get some sort of quantified comparison.

If it turns out that auto pilot is too dangerous to use, although based on 3 accidents that is currently not likely, but if it happens the auto-pilot can be disabled over wireless.

Regards,

[ReformedCharacter]

I can foresee a mass recall because of failings with the automatic pilot software. Several have been reported already. Ultimately there could be prohibition.

TJH

That surprises me but I respect your opinion. I imagine that any issues would be dealt with by a software update rather than a recall as such.

It seems likely that the coming debate about the safety of auto-driven vehicles will be driven as much by emotion as logic. It is my view that eventually only auto-driven vehicles will be allowed on the roads most of the time and that there will be some form of mesh communication system between vehicles.

But anyway in the meantime I wonder what will happen when there is sufficient statistical evidence to make a meaningful comparison between human and auto-driven vehicles. If for example in the UK we estimate 1600 deaths per year from human driven vehicles but 1500 equivalent deaths per year from auto-driven vehicles would we be worrying about being driven to our deaths by our cars, and suing the manufacturers? Or would we be grateful for the safety improvement?

RC

[vrdiver]

If for example in the UK we estimate 1600 deaths per year from human driven vehicles but 1500 equivalent deaths per year from auto-driven vehicles would we be worrying about being driven to our deaths by our cars, and suing the manufacturers? Or would we be grateful for the safety improvement?

RC

I'd expect anything that was statistically close to human error based driving deaths would get banned. If the numbers came out as safer by an order of magnitude, then the argument for accepting auto-driven fatalities becomes much stronger.

In my mind, assuming no sensationalist newspaper campaigns, will be the insurance companies and their assessments of risk. If they start favouring driverless cars or pushing driverless as the acceptable path for inexperienced motorists to be able to afford car ownership, then it will become a done deal.

VRD

[BobbyD]

Since Tesla launched its first car, how many reports have you seen of the deaths in cars made by Ford, GM, GE, Toyota, Volvo ...

Of the ten people who will on average die on UK roads today, how many will involve a Tesla and how many will be report. Hint, if a Tesla isn't involved there will be no report, if a Tesla is it will make the main media channels

All of the crash and injury information is freely available as are the clear and document cases when Toyota cars killed people due to defective throttles that stuck open, but one does not see reports on them. E.g. what is the chance of being killed if the car(s) you own are involved in a crash. That would be interesting, but who bothers to research it.

As yet nobody has been killed by a Toyota because Toyota

has never been one to conform just for the sake of conformance.

and decided to sell a car without brakes. You produce enough of anything and you will produce examples with engineering faults, but that isn't what is killing people in incidents with Teslas, which comprise a minuscule percentage of cars on the road.

Tesla are putting hardware considered inadequate by everybody else in the industry out in to the wild, in a still experimental area, where incorrect implementation is likely to increase risk because the human driver is likely to switch off, and using terms which lead people to overestimate the hardwares capabilities. I don't care what make the car is, somebody gets caught taking a nap in the passenger seat whilst going down the freeway instead of sitting with his hands on the wheel and it's going to make the papers regardless of the make of car. Toyotas develop a habit of seeking out parked cars to crash in to and it's going to make the papers... Toyotas decide to accelerate and head straight for the motorway divides, guess what, its going to make the papers.


You aren't comparing like with like. Your comparison is a categorisation error.

I can foresee a mass recall because of failings with the automatic pilot software. Several have been reported already. Ultimately there could be prohibition.

They've already been forced to settle a law suit regarding auto pilot

NEW YORK (Reuters) - Tesla Inc on Thursday reached an agreement to settle a class action lawsuit with buyers of its Model S and Model X cars who alleged that the company’s assisted-driving Autopilot system was “essentially unusable and demonstrably dangerous.”

- https://www.reuters.com/article/us-tesl ... SKCN1IQ1SH

If it turns out that auto pilot is too dangerous to use, although based on 3 accidents that is currently not likely, but if it happens the auto-pilot can be disabled over wireless.

At which point what happens to pre-orders and company reputation, or the class action law suit that follows?

I can foresee a mass recall because of failings with the automatic pilot software. Several have been reported already. Ultimately there could be prohibition.

TJH

That surprises me but I respect your opinion. I imagine that any issues would be dealt with by a software update rather than a recall as such.

...and if the problem isn't the software, but an insistence on running insufficient hardware for autonomous driving given where autonomous driving sits as an embryonic technology?

[ReformedCharacter]

...and if the problem isn't the software, but an insistence on running insufficient hardware for autonomous driving given where autonomous driving sits as an embryonic technology?

Yes, an entirely different problem indeed.

RC

[odysseus2000]

BobbyD
You aren't comparing like with like. Your comparison is a categorisation error.

Correct, I am comparing the accepted status quo of around 10 UK road deaths per day, to the potential to reduce that number.

We are at a point where we have accepted the road traffic accident and fatality rates as standard and we as a driving species are now experimenting with ways to reduce the number of deaths.

3 People across the world have died in Tesla vehicles under computer control.

Around 3000 people died last year in the UK alone with human drivers.

Sense suggests that if you can find a way to reduce this it would be a good thing.

The question then comes down to cost: Can you make a robot car that is much safer than a human driven and that is affordable?

If you believe that you can, how do you test it?

What ever technology you create to make driving safer it will have to be tasted and likely refined and there will be a cost in blood and treasure as that process evolves.

One would like to get it right first go, but life and the laws of thermodynamic don't work that way.

Hence, either you do this or you accept that 3,000 folk will die this year, the next and on and on.

Regards,

[BobbyD]

BobbyD
You aren't comparing like with like. Your comparison is a categorisation error.

Correct, I am comparing the accepted status quo of around 10 UK road deaths per day, to the potential to reduce that number.

We aren't talking about the lives that might be saved by autonomous driving, we are talking about the people being killed in incidents involving cars with Tesla's overgrown cruise control. Auto pilot is not, and shows no signs of becoming, autonomous driving.

Around 3000 people died last year in the UK alone with human drivers.

All the Tesla accidents also had human drivers. Auto pilot is not autonomous driving.

There are autonomous vehicles out there on the street today. None of them are Teslas.

Around 3000 people died last year in the UK alone with human drivers.

Sense suggests that if you can find a way to reduce this it would be a good thing.

...and autonomous vehicles will be a great step forwards.

These aren't them.

[odysseus2000]

BobbyD
...and autonomous vehicles will be a great step forwards.

These aren't them.

No one knows what autonomous vehicles will look like.

The engineering task is to make a car at sufficiently low cost that is a lot better than a human driver.

The Tesla system is one approach.

There are other much more complicated approaches which have relatively low mileage data.

We can all have opinions but we do need real world data.

Regards,

[BobbyD]

No one knows what autonomous vehicles will look like.

They probably won't look like they just tail ended a stationary police car because their blind spot started at the car in front's rear bumper and when the car pulled out to go around the police car the car decided, hey I'm a driver assistance module not an autopilot, collision avoidance just isn't my thing man.

The engineering task is to make a car at sufficiently low cost that is a lot better than a human driver.

The first thing you need to do for autonomous driving is build an autonomous car. You can improve and streamline and bring in economies of scale later. The impressive bit is building the thing in the first place. In this tiny aspect of the race for autonomous driving Tesla are years behind.

There are other much more complicated approaches which have relatively low mileage data.

...because some people consider taking a system which occupies a dangerous no mans land between machine control and human control, running on insufficient data, putting it in to the hands of real people on real roads and telling them it is an autopilot to be incredibly irresponsible. But hey, health and safety gone mad right?

The Tesla system is one approach.

Taking a sheep brain mounting it on the engine and connecting the occipital lobe to the motor control unit with jump leads is an approach, and stupid as it sounds it's an approach which would have resulted in three fewer deaths. Being an approach isn't an argument.

[odysseus2000]

BobbyD
...because some people consider taking a system which occupies a dangerous no mans land between machine control and human control, running on insufficient data, putting it in to the hands of real people on real roads and telling them it is an autopilot to be incredibly irresponsible. But hey, health and safety gone mad right?

You mean like human drive cars and all the folk they have killed.

There is a point were one looks at the greater good as well as the downside.

Cars have always been major killers, but the politicians have taken the view that the good they do outweighs the deaths they cause.

Regards,