Welcome to the North American Subaru Impreza Owners Club Saturday December 15, 2018
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Go Back   NASIOC > NASIOC General > News & Rumors > Non-Subaru News & Rumors

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads. 
Reply
 
Thread Tools Display Modes
Old 12-01-2018, 08:28 AM   #1
AVANTI R5
Scooby Guru
 
Member#: 73805
Join Date: Nov 2004
Default Ready for self-driving future? Sorry, it's drizzling out.



An autonomous vehicle belonging to Michigan State University parks on a curb in the Spartan Village section of campus, which engineers are using to gather AV-related data and conduct test drives. Photo credit: Pete Bigelow
Quote:
Even a little crummy weather might pose big problems for self-driving systems.

New research from Michigan State University suggests light rain and drizzle can confound the algorithms that autonomous systems use to detect pedestrians, bicyclists and other road users.

The findings raise the prospect that until these algorithms can better handle a variety of weather conditions, self-driving vehicles may be limited to Sun Belt states, or fleets of vehicles might need to be grounded when weather conditions are subpar.

"When we run these algorithms, we see very noticeable, tangible degradation in detection," said Hayder Radha, an MSU professor of electrical and computer engineering who oversaw the study. "Even low-intensity rain can really create some serious problems, and as you increase the intensity, the performance of what we consider state-of-the-art mechanisms can almost become paralyzed."

The researchers are finalizing their report, but Radha previewed the findings in a conversation with Automotive News.

Although radar and lidar are often used to detect obstacles, Radha said the research focuses on measuring the competence of computer vision systems because cameras are most often the primary sensor automakers and tech companies use to classify pedestrians and other road users.

But the problem is not the cameras, Radha stressed; it's the algorithms distilling information from them.

"Once you throw in a few drops of rain, they get confused," he said. "It's like putting eyedrops in your eye and expecting to see right away."

Cold reality

Researchers evaluated various parameters for the study, including the size of the raindrops, the number of raindrops per square inch and the effect of wind velocity. Using a scale that ran from clear weather to a blinding rainstorm, they found algorithms failed to detect as many as 20 percent of objects when the rain intensity was 10 percent of the worst-case scenario. When rain intensity increased to 30 percent, as many as 40 percent of objects could no longer be detected.

Information from radar and lidar may mitigate some of the problems, especially as self-driving engineers develop ways to use returns from those sensors to classify objects. But the most improvement will come, Radha says, when self-driving tech is routinely tested beyond sunny locations in Arizona, Florida and Texas.

The ongoing research is one of Michigan State's efforts to study self-driving technology and mobility. The school created an interdisciplinary program called CANVAS Connected and Autonomous Networked Vehicles for Active Safety part of which examines the situational-awareness capabilities of self-driving systems, including their competence in detecting and handling adverse weather.

Broadly, Michigan State has strived to increase its focus on autonomous systems and advanced transportation.

That has been an ambition throughout the state's higher-education ranks. Initially, the University of Michigan was a primary focus thanks to its Mcity closed-course autonomous-vehicle test track. But more recently, the state has sought to ensure all its schools are keeping mobility in mind. Last year, Washtenaw Community College in Ann Arbor launched an Advanced Transportation Center to help train field technicians to build and repair intelligent transportation systems.

At Michigan State, researchers have affixed cameras to traffic lights around the university's home in East Lansing, allowing them to gather data on pedestrian, bicyclist and vehicle movements and better predict their movements, a crucial capability for self-driving systems. Elsewhere on campus, university officials have reclaimed Spartan Village, an area filled with dilapidated apartments, and used it as a public course to test autonomous technology and gather data.

Seasonal effect

That testing has yielded another weather-related finding. Depending on the environment, high-resolution maps that autonomous systems use to determine their location may need seasonal updates. As the leaves fell from towering trees in Spartan Village, researchers realized their maps were outdated.

"You can imagine in environments where there are a lot of leaves on trees or on shrubs close to the road, they are an essential part of the map," Radha said. "So summer and winter are completely different. When they fall down in winter, you have nothing to work with. So that tells you that for this technology to be robust, it needs to be developed in different conditions than you see only in Arizona and Silicon Valley."

Cold weather can also cause problems for autonomous vehicles. Last winter, he found temperatures of 10 degrees Fahrenheit and lower increased the "noise," or amount of poor-quality or irrelevant returns from lidar sensors. He says manufacturers told him they could not guarantee the sensors would operate in extreme cold, and that they're working to expand the range of the sensors.

Whether it's cold weather, foliage or light rain, Radha says the shortcomings underscore the fact that weather diminishes the capabilities of self-driving technology for the foreseeable future.

"Frankly, at first we thought we'd look at this for a couple of years and then be done," he said. "But I'd say now that challenging weather conditions are going to be a problem for many years to come."
https://www.lansingstatejournal.com/...her/931651001/
* Registered users of the site do not see these ads.
AVANTI R5 is offline   Reply With Quote
Sponsored Links
* Registered users of the site do not see these ads.
Old 12-01-2018, 09:59 AM   #2
Integra96
Scooby Specialist
 
Member#: 64196
Join Date: Jun 2004
Chapter/Region: RMIC
Vehicle:
2010 Family Sedan

Default

Problems are soluble. This one, particularly so.
Integra96 is offline   Reply With Quote
Old 12-01-2018, 06:19 PM   #3
JustyWRC
Scooby Specialist
 
Member#: 153088
Join Date: Jul 2007
Location: Arlington, TN
Vehicle:
2005 Baja Turbo
06/04 For 18 Imp 4dr Sprt

Default

Just like most anything, it's just a matter of time.....
JustyWRC is offline   Reply With Quote
Old 12-03-2018, 09:02 AM   #4
Sid03SVT
Scooby Specialist
 
Member#: 183032
Join Date: Jun 2008
Location: CT
Vehicle:
OLD Family Wagon

Default

Quote:
Originally Posted by AVANTI R5 View Post

Whether it's cold weather, foliage or light rain, Radha says the shortcomings underscore the fact that weather diminishes the capabilities of self-driving technology for the foreseeable future.
Yeah the average driver can't seem to handle these conditions either, but hey, we can ignore that fact I guess.

in ten years +/- this technology will have improved significantly, while the first snowflake that falls will still be responsible for thousands of CUV's to slide off the road in New England because it's the first time it's snowed here, ever, again.
Sid03SVT is offline   Reply With Quote
Old 12-03-2018, 10:53 AM   #5
Scooby921
Merci Buckets
Moderator
 
Member#: 88606
Join Date: Jun 2005
Chapter/Region: MWSOC
Location: Michigan
Vehicle:
2016 Cayman GTS
Dark Blue Metallic

Default

Quote:
Originally Posted by Sid03SVT View Post
Yeah the average driver can't seem to handle these conditions either, but hey, we can ignore that fact I guess.

in ten years +/- this technology will have improved significantly, while the first snowflake that falls will still be responsible for thousands of CUV's to slide off the road in New England because it's the first time it's snowed here, ever, again.
People are going to lose their **** when fully autonomous cars are finally viable and a ****box will have a $100k price tag because of all the technology required to make it work.
Scooby921 is offline   Reply With Quote
Old 12-03-2018, 11:48 AM   #6
SCRAPPYDO
Scooby Guru
 
Member#: 873
Join Date: Feb 2000
Chapter/Region: TXIC
Location: www.testdrivemylife.com
Vehicle:
2018 Leg / Ridgeline
Datsun 71 240Z & 68 2000

Default

Quote:
Originally Posted by Scooby921 View Post
People are going to lose their **** when fully autonomous cars are finally viable and a ****box will have a $100k price tag because of all the technology required to make it work.
so much this. They are in such a damn hurry to make this work, but never stopped to think if anybody actually WANTS it and wants to pay for it. Instead of doing all this autonomous crap, why not just make cars that do not suck.
SCRAPPYDO is offline   Reply With Quote
Old 12-03-2018, 12:05 PM   #7
godfather2112
Papi Chulo
Moderator
 
Member#: 53794
Join Date: Jan 2004
Chapter/Region: RMIC
Location: Boner kill city
Vehicle:
... White Magic
Jesus Underwear

Default

Quote:
Originally Posted by SCRAPPYDO View Post
so much this. They are in such a damn hurry to make this work, but never stopped to think if anybody actually WANTS it and wants to pay for it. Instead of doing all this autonomous crap, why not just make cars that do not suck.
Waymo is doing this correctly. While yes, its all about how fast can you get the product / service to the customer, they are not rushing it nor should they. The one thing I think will be interesting is how they autonomous cars drive during severe weather conditions (i.e how much do they slow down the cars speed). It was dumping snow here the last few days and you had people who were doing maybe 40mph on the freeway to cars and trucks doing 70mph without issue.
godfather2112 is online now   Reply With Quote
Old 12-03-2018, 12:06 PM   #8
Masterauto
Scooby Specialist
 
Member#: 198376
Join Date: Dec 2008
Chapter/Region: MAIC
Location: Delaware
Vehicle:
17 Alfa GiuliaTi S
Nissan GT-R 15, STI 15

Default

they will have to digitally mark all other vehicles, road features, bicyclists etc
the current Waymo visual system will not work heavy rain, fog, snow etc.
Its going to take lots longer than they say on media.
Tesla stock buyers look into Robosense https://www.robosense.ai/news

Last edited by Masterauto; 12-03-2018 at 12:30 PM.
Masterauto is offline   Reply With Quote
Old 12-03-2018, 12:41 PM   #9
Scooby921
Merci Buckets
Moderator
 
Member#: 88606
Join Date: Jun 2005
Chapter/Region: MWSOC
Location: Michigan
Vehicle:
2016 Cayman GTS
Dark Blue Metallic

Default

Quote:
Originally Posted by godfather2112 View Post
Waymo is doing this correctly. While yes, its all about how fast can you get the product / service to the customer, they are not rushing it nor should they. The one thing I think will be interesting is how they autonomous cars drive during severe weather conditions (i.e how much do they slow down the cars speed). It was dumping snow here the last few days and you had people who were doing maybe 40mph on the freeway to cars and trucks doing 70mph without issue.
Honestly, everyone other than Tesla is doing it correctly. Tesla put Autopilot to market to say they did it first, and then blame the consumer any time there is an issue. All of the major OEM's and plenty of startups have been working on autonomous development for much longer and haven't rushed anything to market because they want to do it right the first time. I suspect Tesla's issues (customers's issues) with Autopilot is all the more incentive for others to take their time.
Scooby921 is offline   Reply With Quote
Old 12-03-2018, 12:55 PM   #10
Integra96
Scooby Specialist
 
Member#: 64196
Join Date: Jun 2004
Chapter/Region: RMIC
Vehicle:
2010 Family Sedan

Default

Quote:
Originally Posted by Scooby921 View Post
People are going to lose their **** when fully autonomous cars are finally viable and a ****box will have a $100k price tag because of all the technology required to make it work.
Right, because that's how technology economics works - everything just gets more expensive with widespread adoption.
Integra96 is offline   Reply With Quote
Old 12-03-2018, 12:55 PM   #11
SCRAPPYDO
Scooby Guru
 
Member#: 873
Join Date: Feb 2000
Chapter/Region: TXIC
Location: www.testdrivemylife.com
Vehicle:
2018 Leg / Ridgeline
Datsun 71 240Z & 68 2000

Default

I would think you would need some kind of road impregnated with magnets or sensors, or something passive that the car could pick up on. Maybe on dedicated 'autonomous roadways' they can do this and slowly get it to market. I think putting a 21st century car on a 19th century motorway is problematic.
SCRAPPYDO is offline   Reply With Quote
Old 12-03-2018, 01:04 PM   #12
SirBrass
Scooby Specialist
 
Member#: 207263
Join Date: Apr 2009
Chapter/Region: Tri-State
Location: Northern PA
Vehicle:
2019 WRX Limited
World Rally Blue

Default

Quote:
Originally Posted by SCRAPPYDO View Post
so much this. They are in such a damn hurry to make this work, but never stopped to think if anybody actually WANTS it and wants to pay for it. Instead of doing all this autonomous crap, why not just make cars that do not suck.
The only people who want it, really, are people who are otherwise taking public transportation already. They just want the same "get me there and don't bug me with the driving" benefit without having to take a greyhound bus in a personal vehicle for longer trips.

Sorry, but not going to happen even when viable. A human command presence is always going to be required for safety, and an alert, attentive and in control human is always faster to react to a computer screw up than an inattentive, constantly distracted human who doesn't want to be the driver anyway.

Cost will always go down as the technology matures and manufacturing and sourcing gets ramped up for production, so that's a short term problem really.

It's more the philosophy of the thing that is the major issue for me. Self-driving cars are dangerous, especially since their target market are folks who don't want to be bothered with operating it anyway.

Aircraft under autopilot are different, as you have trained crew who wanted to do this keeping an eye on things and also there's more of a "damn the cost, this has to be reliable beyond belief" factor in the autopilot avionics as well. That, and the folks who own that aircraft are 99% of the time employing the folks flying it, too, so the pilots aren't forking over the millions that the aircraft cost to purchase.

Not to mention that autopilot, IIRC, cannot be used in congested airways, per FAA rules. Apply that logic to the roadways, and 90% of where folks want these self-driving cars would not be allowed to use the self-driving functions.

So, the massively expensive and far more robust and mature autopilot technologies for autonomous aircraft operation can't reliably do what we want to fit into a much smaller passenger vehicle, operating with much less margin for time-to-correct-for-error, and with a significant "Who wants to pay that much for it?" factor.

The only place this stuff belongs is in highway cruising.

That, or have central traffic computers linking commuters into "commute trains" and operating them through traffic like multiple trains on rails. You pick which "train route" you want, go to the nearest "access station", and the traffic computers hopefully link you up with one flowing nearby or use you to start one.

That's the really advanced sci-fi future vision, I think, and it basically is just a more flexible version of already existing public transportation without the freedom of movement that a personal automobile is supposed to represent.

Thus, screw that noise.

These projects are solidly falling into Ian Malcom's warning about "should" and "could."
SirBrass is offline   Reply With Quote
Old 12-03-2018, 01:06 PM   #13
SCRAPPYDO
Scooby Guru
 
Member#: 873
Join Date: Feb 2000
Chapter/Region: TXIC
Location: www.testdrivemylife.com
Vehicle:
2018 Leg / Ridgeline
Datsun 71 240Z & 68 2000

Default

Ian was where I got my statement from...
SCRAPPYDO is offline   Reply With Quote
Old 12-03-2018, 01:06 PM   #14
VaporTrail308
Scooby Newbie
 
Member#: 373439
Join Date: Nov 2013
Default

I think people grossly underestimate the requirements to make self driving cars a reality on a mass scale.

Suffice to say that my business has done a lot with automation in manufacturing and it's not nearly as simple as people think it is. And that's in a controlled, repetitive environment. Add in a totally uncontrolled, non-repetitive environment and that's a whole new level of complexity. I think it's a lot further off than most people think.
VaporTrail308 is offline   Reply With Quote
Old 12-03-2018, 01:22 PM   #15
SirBrass
Scooby Specialist
 
Member#: 207263
Join Date: Apr 2009
Chapter/Region: Tri-State
Location: Northern PA
Vehicle:
2019 WRX Limited
World Rally Blue

Default

Quote:
Originally Posted by VaporTrail308 View Post
I think people grossly underestimate the requirements to make self driving cars a reality on a mass scale.

Suffice to say that my business has done a lot with automation in manufacturing and it's not nearly as simple as people think it is. And that's in a controlled, repetitive environment. Add in a totally uncontrolled, non-repetitive environment and that's a whole new level of complexity. I think it's a lot further off than most people think.
Exactly. There's a reason machinists and electronics soldering techs can make so much bank, still. The automation systems and robots still screw things up or can't do things safely (to the item under manufacture) that you still need highly skilled humans supervising and doing other tasks and with their hands on the big red button.

As you said, that's in a controlled environment.

I work in the avionics industry as a systems engineer. There's a reason aircraft don't tie in collision avoidance to the autopilot, and why the industry is just starting to implement that in a very limited capacity into UAVs. That reason is, you need the pilots to filter out things which a human can quickly discern are false positives, but which would be massively complex and bug-ridden if you try and program a computer to do the same.

It's a reliability issue rather than a capability issue. You need the human's ability to interpret and interpolate and deduce, which machines simply can't do or can do only very badly, to operate in these environments.

I was driving up to Buffalo last night and encountered severe ground fog. It was very difficult to see the road. I had to switch to my nav app on AA so I could see the road route while I concentrated on catching glimpses here and there of the center and side lines. I knew they were there, but a computer wouldn't. Not with the cameras getting their lenses coated in condensation, and not being able to penetrate the fog very well.

Computers simply do not see what we clearly see or see it as well.

Ever had trouble getting a good picture that isn't too dark or too bright even with auto-iso, even though your two eyes see it just fine with proper contrast and no wash-out, but the camera can't get a good image for love nor money?

Yeah, that same tech is being relied on in cars to see things and stop or avoid with you being distracted. Or at least that's what folks want.

We simply do not appreciate the many orders of magnitude of automatic processing and decision making that we do in our brains to accomplish the task of driving a car safely, or even just walking. And we overestimate by orders of magnitude the capability of cheap imitations of human sensory organs (especially the eyes) to inform cheap imitations of human brains (computers) which insult the mental processing of even the most developmentally retarded individual. Sure they can do things faster than we can, but that's b/c they are stupid and given very simple tasks that are very highly controlled and very very easily repeatable. But when it comes to making sense of the patterns in a chaotic environment, they aren't up to the task, when our human brains and eyes do it very easily.

Quote:
Originally Posted by SCRAPPYDO View Post
Ian was where I got my statement from...
I know, that's why I thought of it. It was "Hey, that sounds familiar...." I was agreeing with you by adding my own take.
SirBrass is offline   Reply With Quote
Old 12-03-2018, 01:34 PM   #16
Kevin
Scooby Specialist
 
Member#: 1082
Join Date: Mar 2000
Chapter/Region: VIC
Location: Vancouver, BC
Vehicle:
2017 X3
Old Man Silver

Default

Autopilot and self driving have completely different challenges IMO. With autopilot, once they figured out reliable control of aircraft the environmental interaction was easy. With self driving, control of the vehicle is easy but interaction with the surrounding environment must be immensely difficult to deal with.
Kevin is offline   Reply With Quote
Old 12-03-2018, 02:14 PM   #17
dwf137
Scooby Specialist
 
Member#: 161333
Join Date: Oct 2007
Chapter/Region: NWIC
Location: Seattle
Vehicle:
'18 Golf R
'14 Outback

Default

Quote:
Originally Posted by Scooby921 View Post
People are going to lose their **** when fully autonomous cars are finally viable and a ****box will have a $100k price tag because of all the technology required to make it work.
People (aside from the type already in 100K plus cars) will not be buying them. Companies will buy them and put them on the road to replace drivers. After all, a 100K **itbox will be immensely cheaper than paying a person.

Because isn't that the goal? Automate all of the jobs low- and middle-class folks currently do so the richies can stop paying them... then the richies get to complain when the people they just made unemployed need help.
dwf137 is offline   Reply With Quote
Old 12-03-2018, 03:43 PM   #18
SCRAPPYDO
Scooby Guru
 
Member#: 873
Join Date: Feb 2000
Chapter/Region: TXIC
Location: www.testdrivemylife.com
Vehicle:
2018 Leg / Ridgeline
Datsun 71 240Z & 68 2000

Default

That is quite a reach their buddy..
SCRAPPYDO is offline   Reply With Quote
Old 12-03-2018, 04:03 PM   #19
SirBrass
Scooby Specialist
 
Member#: 207263
Join Date: Apr 2009
Chapter/Region: Tri-State
Location: Northern PA
Vehicle:
2019 WRX Limited
World Rally Blue

Default

Quote:
Originally Posted by dwf137 View Post
People (aside from the type already in 100K plus cars) will not be buying them. Companies will buy them and put them on the road to replace drivers. After all, a 100K **itbox will be immensely cheaper than paying a person.

Because isn't that the goal? Automate all of the jobs low- and middle-class folks currently do so the richies can stop paying them... then the richies get to complain when the people they just made unemployed need help.
That's.... yeah, no. Your conclusion is wrong and your reasoning incredibly off-base.

Quote:
Originally Posted by Kevin View Post
Autopilot and self driving have completely different challenges IMO. With autopilot, once they figured out reliable control of aircraft the environmental interaction was easy. With self driving, control of the vehicle is easy but interaction with the surrounding environment must be immensely difficult to deal with.
What I'm saying is that aircraft already have instrumentation and electronics that could fairly easily concept-wise be integrated with autopilot to do things that we want self-driving cars to do, which is avoid hitting things.

There's a reason that they do not do that, and that reason should show why its much harder especially for the cost desired to integrate that into automobiles, and that's also not factoring in the truth that the aerospace environment is an easier environment to operate in. Spacing is in miles, and you generally always have seconds at least within which to react, unlike in the car where in an emergency you may have only very few seconds to only a split second to see, comprehend, and react. Humans can sometimes be slow on the reacting thing, but right now computers are really bad comparatively at the "seeing and comprehending" part. When it's bloody obvious to everyone and their blind mother, computers are much faster than humans (and even moreso distracted humans, aka your typical California driver).

In an aircraft, the electronics help because they can see farther than humans can and at wider angles (TCAS monitors omnidirectionally for transponder signals, for instance, where the pilots only have a limited field of view out of the windscreen), and thus give the pilots lots of warning (often minutes and minutes). Now, ground avoidance is a bit more urgent, but still, it's not linked to autopilot for safety reasons.

Which should be telling everyone that they're going to need computing tech that is just as dead nuts reliable as flight safety avionics are required to be but much much more reliable at processing and interpreting input than avionics safety equipment, at a fraction of the cost added to the price of the vehicle.

We're going to need more development, lots more development, in this area before that's even remotely viable, because right now it sure as hell is not. When the sensors are accurate enough, the computers regularly glitch out, and when the system is reasonably buffered against false positives they're then incredibly prone to false-negatives as well and thus become more unsafe than a distracted driver in traffic to everyone around them.

Last edited by SirBrass; 12-03-2018 at 04:13 PM.
SirBrass is offline   Reply With Quote
Old 12-04-2018, 12:03 PM   #20
justincredible
Scooby Specialist
 
Member#: 83633
Join Date: Mar 2005
Chapter/Region: RMIC
Location: The Mountains
Vehicle:
2012 bright green
box

Default

There were 250 wrecks yesterday morning because it snowed... in Utah... like it does every year.

Autonomous cars can't be any worse.
justincredible is offline   Reply With Quote
Old 12-04-2018, 12:50 PM   #21
godfather2112
Papi Chulo
Moderator
 
Member#: 53794
Join Date: Jan 2004
Chapter/Region: RMIC
Location: Boner kill city
Vehicle:
... White Magic
Jesus Underwear

Default

Quote:
Originally Posted by justincredible View Post
There were 250 wrecks yesterday morning because it snowed... in Utah... like it does every year.

Autonomous cars can't be any worse.
Autonomous cars would need to be able to correct a slide, and choose which way to go to avoid a crash. If you see cars up ahead sliding around, it could give you more time to prepare and slow down versus autonomous. Obviously in the future all the cars should communicate with each other and prevent the above stated.
godfather2112 is online now   Reply With Quote
Old 12-04-2018, 01:05 PM   #22
SirBrass
Scooby Specialist
 
Member#: 207263
Join Date: Apr 2009
Chapter/Region: Tri-State
Location: Northern PA
Vehicle:
2019 WRX Limited
World Rally Blue

Default

Quote:
Originally Posted by justincredible View Post
There were 250 wrecks yesterday morning because it snowed... in Utah... like it does every year.

Autonomous cars can't be any worse.
You want to bet? I'd be more than happy to take your money .
SirBrass is offline   Reply With Quote
Old 12-04-2018, 02:21 PM   #23
torquemada
Scooby Guru
 
Member#: 128484
Join Date: Oct 2006
Chapter/Region: International
Location: Germany
Vehicle:
2006 EDM WRX STI
WRB

Default

Quote:
Police Pull Over Self-Driving Tesla with Sleeping Man Behind the Wheel

There may come a day when cars can truly drive themselves and you can take a nap during your commute, but we’re not there yet. In the meantime, the ever-advancing state of self-driving technology has made some people a bit too comfortable turning over control to a system that can’t respond as effectively as a human driver. The California Highway Patrol recently pulled over a Tesla with some difficulty after police spotted the driver passed out behind the wheel.

The incident took place on Highway 101 when a patrol car spotted a Tesla Model S with a driver that appeared to be asleep. While the Tesla Autopilot system can handle some basic driving tasks, it’s not good enough that you can take a nap. The driver, 45-year-old Alexander Samek, was drunk and had passed out on his way home. You could argue that having the Autopilot system driving was probably safer than giving the heavily impaired human control of the car, but neither situation is legal.

Autonomous driving systems generally fall into one of five levels. Level one includes basic automation like lane assistance. At level two, a car can support for one or more tasks without the driver’s constant interaction — for example, many modern cars can steer and brake for several seconds on the highway without your hands on the wheel. Level three is where most research is currently focused. These vehicles use advanced sensors to scan the environment and drive for extended periods while responding to changing conditions. At level four and five, cars can drive well enough that you don’t have to pay attention at all. Level four can handle most types of driving, and level five is full automation that you never need to think about.

Tesla’s Autopilot system is somewhere between level two and three, so it’s certainly not good enough to take a nap behind the wheel. Patrol cars following Samek’s Tesla were unable to rouse him from his alcohol-fueled slumber, which made stopping the Tesla rather tricky. They eventually worked out a method to use the car’s own sensors to stop it. One patrol car stayed behind the Tesla, driving in a sweeping S-curve to keep other cars from getting in the way. Meanwhile, another cruiser maneuvered in front of the Model S and gradually slowed down. The Tesla’s radar saw a slower car in front, so it too slowed down. Eventually, it stopped in the middle of the highway.

It took about seven miles for police to bring the Model S to a stop. Samek was promptly arrested for drunk driving, but a cunning lawyer might attempt to point out he wasn’t technically driving. Whatever the outcome, this is a brave new world.
https://www.extremetech.com/extreme/...hind-the-wheel
torquemada is offline   Reply With Quote
Old 12-04-2018, 03:14 PM   #24
SirBrass
Scooby Specialist
 
Member#: 207263
Join Date: Apr 2009
Chapter/Region: Tri-State
Location: Northern PA
Vehicle:
2019 WRX Limited
World Rally Blue

Default

Quote:
Originally Posted by torquemada View Post
Clever lawyer indeed. Doubt it'll get the guy off entirely, but a particularly amused judge might not throw the entire book at him, reasoning that the guy's choice to use autopilot, though not preventing him from violating the law and the car's system not sophisticated enough to safely drive was still reasonably assumed to be safer and showed an attempt, though a very impaired attempt, to operate the vehicle safely.

Maybe fine and suspended license instead of losing his license?

Who knows. But the "wasn't technically driving" bit won't actually fly and any decent lawyer knows that.
SirBrass is offline   Reply With Quote
Old 12-04-2018, 08:46 PM   #25
dwf137
Scooby Specialist
 
Member#: 161333
Join Date: Oct 2007
Chapter/Region: NWIC
Location: Seattle
Vehicle:
'18 Golf R
'14 Outback

Default

Quote:
Originally Posted by SirBrass View Post
That's.... yeah, no. Your conclusion is wrong and your reasoning incredibly off-base.
Sarcasm, bro.

but that is the end result.
dwf137 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

All times are GMT -4. The time now is 08:06 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2018 Axivo Inc.
Copyright ©1999 - 2017, North American Subaru Impreza Owners Club, Inc.