Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×

Comment Re:Darwin was right. (Score 1) 329

From a CNN report:

A Tesla spokesperson released a moment by moment description of what happened in the 40 seconds before the crash.

After 15 seconds of what was described by Tesla as "visual warnings and audible tones," the autopilot began to disengage because the driver's hands were still not on the wheel.

About 25 seconds before the crash, "Autosteer began a graceful abort procedure in which the music is muted, the vehicle begins to slow and the driver is instructed both visually and audibly to place their hands on the wheel," according to the company.

Tesla said the driver responded 11 seconds before the crash by retaking the wheel, turning it toward the left and pressing on the accelerator.

"Over 10 seconds and approximately 300m later and while under manual steering control, the driver drifted out of the lane, collided with a barrier, overcorrected, crossed both lanes of the highway, struck a median barrier, and rolled the vehicle," according to Tesla's account.

So, the man never made the decision to disable autopilot. Instead, the car turned off the autopilot on its own. So, Musk could say with a straight face that autopilot was off. But how can a safety mechanism be allowed to turn itself off under any circumstances? Talk about the exact opposite of fail-safe.

Comment Re:Bot manufacturer's press release here: (Score 1) 255

So, when the robot detected a moving child, why TF didn't it just stop? Kids that age move unpredictably when faced with the unexpected. I always stop in circumstances like that. A toddler can dodge a stationary obstacle better than I can dodge a dodging toddler..

The indications from the robot company are that the robot records indicate that it did stop. The child's parent said that the robot ran over the boy but haven't actually claimed to have seen the incident. The robot is designed to move at 1 mph, so it is not designed to dodge anything but rather to stop, which is apparently did. An adult can be expected to dodge a 300-lb robots. A 16-month old toddler can be expected to do unexpected and irrational things, like running backwards into a large robot.

I already figured that the robot didn't sense that it was traveling over the kid's foot, or it would have stopped, so a record of the robot's sense impressions is not all that useful. The injuries to the child are a better guide to what actually happened, since it appears there's confusion there.

The indications are the robot stopped. The injuries to the child are not necessarily indicative of an out-of-control robot. Any parent of a small child can realize how common it is for a small child to acquire injuries and bruises when playing with stationary objects.

Comment Re:Bot manufacturer's press release here: (Score 4, Interesting) 255

The mom said, "The robot hit my son's head and he fell down - facing down - on the floor, and the robot did not stop and it kept moving forward." This is in direct contrast to what the robot company said, so one of the accounts is not accurate.

The robot company also said, "The machine veered to the left to avoid the child, but the child ran backwards directly into the front quarter of the machine, at which point the machine stopped and the child fell on the ground." To make a statement about the orientation of the boy requires video (or at least some other electronic detection). Furthermore, the company said, "The machine’s sensors registered no vibration alert and the machine motors did not fault as they would when encountering an obstacle." So, there is some form of an electronic record of what the robot sensed.

Did the parents or any other human claim to have seen the moment of impact? I don't read any direct claim of an eyewitness account.

Comment Re:Slippery slope? (Score 1) 297

I do drive a Tesla Model S (S90D) with autopilot - EVERY time you engage it, it displays a warning on the dash display to keep your hands on the wheel, plus if there is ever an issue/ambiguity (it works by actually seeing the lines on both sides of the car....so if one of those lines is faded or missing, the car with either get confused/drift or completely disengage autopilot with an audible warning) ANOTHER display will pop up telling you to keep your hands on the wheel.

It's entirely possible that the average Tesla driver pays greater attention to warnings compared to the general population. However, from looking my own personal experience as well as those people that I have driven with, start-up warnings are basically useless and exist mainly to attempt to protect manufacturers from lawsuits. Real-time pop-up warnings are probably more effective but must be very conservative to avoid false positives. This looks like the case with these Tesla accidents. Was there a pop-up warning for the Harry Potter guy?

If safe autopilot operation really requires keeping one's hands on the steering wheel, there there should be an active system to enforce that requirement.

Comment Re:Slippery slope? (Score 2) 297

Funny thing, Autopilot is what this is.

You may have the idea that "Autopilot" means the plane flies itself. Nope. Typically autopilot on the plane means it will fly straight and level until ordered otherwise. The autopilot on a plane absolutely will fly straight into another plane even, the human pilot is expected to take care of that sort of thing.

Technically autopilot implementations on airplanes do exactly what you said and require a measure of continued vigilance on the part of the pilots. However, that is not what the term means in common language. The English idiom of putting something on autopilot means that something will work without any continued vigilance. In a way, this is a brilliant marketing strategy. The term connotes self-driving while denoting strictly not self-driving. A perfect marketing term.

Comment Re:Slippery slope? (Score 1) 297

I don't drive a Tesla either.

According to this review, they are far better than the competition.

As far as I understand, you cannot miss the warning. It's not like an EULA with walls and walls of text.

I imagine it's like the warning on all dedicated GPS systems. At the start, you have to hit a button and maybe also wait a few seconds to ostensibly read the text. I ignore it every time and occasionally violate the warning by hitting a button while driving. My guess is that the Tesla warning has the same efficacy.

Comment Re:Slippery slope? (Score 4, Interesting) 297

When one attempts to make something idiot-proof, nature builds a better idiot. Not necessarily true, but we live in world where innovators are hampered by the chance of being sued by idiots who just-don't-listen.

"Fire is hot", "peanuts may contain peanuts", "online play not rated", "cruise control is not auto-pilot", "autopilot is experimental", etc.

I don't drive a Tesla, but the only message I heard about Tesla's Autopilot was the name. Yes, there are safety warnings in the manual and when you start up the car, but who actually pays attention that that? The same people who read EULAs? There's a reason the product is called Autopilot and not assist or level-2, and the reason is that they want to implicitly convey the idea that they are better than the competitors with mere assist or level-2. The name is not accidental.

Comment Re:So twice as safe then? (Score 1) 379

You realize that traffic fatalities are a multiple-times-daily occurrence in the USA alone, right? That's not some fuzzy guesstimate, it's about as statically sound as you could hope for. 94M miles (the number Tesla gives per fatal accident in the US, which is a better comparison than the idiot submitter and CNBC author chose to display) is nothing in a country with over 2.5 times that many vehicles.

This is a matter of statistics and sampling. What I'm wondering about is the statistical confidence interval for the stated numbers. It's not at all clear that the confidence intervals are non-overlapping and that they would pass a statistical hypothesis test. Are you aware of additional data that would allow such hypothesis testing?

But more importantly, I conjecture that the underlying populations being sampled are significantly different. For example, my guess is that there are very few if any under-25 Tesla drivers and that such drivers account for a disproportionate amount of the fatalities in the general population. There are probably other such demographics that are prone to fatal accidents that are not representative of the Tesla population.

Comment Re:So twice as safe then? (Score 1) 379

That's still pretty impressive if it's twice as safe as letting a human drive.

Even more so after seeing all the videos on youtube with people in the back of the car letting tesla drive.

It's not clear to me that the Tesla system is safer than a human based on the quoted numbers. First, the incidents are very unlikely for either human or Tesla, so it's not clear that once in 130e6 or once in 60e6 miles is statistically different. Second, the populations for the two numbers are definitely different. Tesla owners are clearly not representative of the general population, so the more apples-to-apples comparison is between Autopilot and manual driving for the same type of drivers, probably characterized by income, education, age, race, tendency to drive/party late at night, tendency for drunken driving, etc.

Comment Re:The USA is Huge (Score 1) 136

The UK is pretty small.

Yeah and the US is large. But New York city has a higher population density than Tokyo, yet only a fraction of the internet speeds. So while you can argue that there is a large area with no or low speed access, you can't excuse crap service in prime areas

It would be really interesting to see the distribution of ISP speeds by country. For example, the Akamai report shows that while South Korea's average speed is 29.0 Mbps, the majority (58%) of South Korean connections are slower than 25 Mbps. Thus, there are perhaps some very high speed connections that somewhat inflate the average. Absent some representation of the distribution, perhaps the median speed would be more representative.

It's also interesting to note that the study methodology is significant. For example, Akamai's methodology and results starkly differ from the FCC's findings, which estimates the average speed in the US at 31 Mbps in 2014 and quickly increasing.

The distribution of speeds is dependent not only on technical infrastructure but also on pricing. A very large percentage of Americans have very high speed internet access offered in their area, but many may choose lower speeds due to economic affordability or the realization that higher speeds are not needed due to personal usage patterns or the existence of speed bottlenecks aside from the ISP (e.g., internet servers, multimedia protocols, etc.).

Comment Re:Unsurprising (Score 1) 441

Billboards and other sources of ambiguous sensory input will be banned from roads.

When drivers don't have to look at their own dashboard, they are more likely to look at billboards and other sources of ambiguous sensory input.

You mean, there will be no need to ban billboards because no advertiser will rent them.

Maybe some low-end car manufacturer will eventually find a way to pipe ads into a car in exchange for a lower upfront purchase price. Then the car becomes an advertising prison.

And garbage along highways will not be noticeable and therefore will no longer be illegal ... well ...

Comment Re:A preview of President Trump's upcoming win. (Score 4, Insightful) 693

the third-worlder isn't all that much better off than before, and may actually be much worse off if they went from an agricultural job they had some control over their destiny to a dismal factory job where they have no control at all

This is a first world perspective. Control and quality of life are not concerns until subsistence is no longer a concern.

Slashdot Top Deals

Everybody likes a kidder, but nobody lends him money. -- Arthur Miller

Working...