erikem NorthWest What if the car suddenly gives back control when an icy skid becomes outside parameters? All the more reason to keep the driver informed of what the car was doing, in particular feedback through the steering wheel. A co-worker suggested that the "auto-pilot" mode should do what's done on locomotives, have an alerter function that requires interaction from the driver at intervals to check if the driver is paying attention.
NorthWest
What if the car suddenly gives back control when an icy skid becomes outside parameters?
All the more reason to keep the driver informed of what the car was doing, in particular feedback through the steering wheel.
A co-worker suggested that the "auto-pilot" mode should do what's done on locomotives, have an alerter function that requires interaction from the driver at intervals to check if the driver is paying attention.
Every single person I know in human factors engineering, IxD and artificial consciousness thinks that the premise and execution of alerters is pointless and, basically, more dangerous than effective. And is prepared to back that up with evidence.
The way to 'check that the driver is paying attention' is to monitor their attentiveness in the background, and periodically interact with them through 'normal' methods (such as conversation). This permits some rather simple confirming analytics, such as voice stress analysis, as well as confirming the right level of 'high functioning' that is necessary for safe response to "emergent situations" (which is how we now have to redefine 'emergencies' with the original word having lost that technical meaning).
An issue with skids is that feedback 'through the wheel' may be inadequate or wrong for many drivers, who will overcompensate or just plain freeze when presented with it. The same has been true of antilock brakes since the early days of hydraulic servomotor actuation, where the default 'advice' to the uninitiated has been 'stomp and steer' even as some of the instantiations have made that response deadly under what may be fairly common circumstances. (It happened directly to me, so I know the issue quite well.) Furthermore, there is little chance that a sufficiently 'total' failure of automatics to cope with a given situation can be addressed by any human response -- nominally "prepared to take over" or not. That is especially true in the '60s-style vehicles made popular by people like the illustrious Syd Mead, where they'd have to drop their drinks and their pinochle hands, turn their chairs 90 degrees, and scramble for pop-out controls ... good luck with that. But it may be true for skids as well; by the time the operator is responding, it may take the skills and reflexes of a Bondurant pursuit graduate just to make the resulting impact 'safe' enough to preserve the lives of the passengers.
There is a very wide set of research results on the psychophysics of 'anticipated catastrophe' that indicate that attempting to keep an operator 'ready' to assume the functions of an automated system is actually worse than not having the automation running under "autonomous control" in the first place. Just imagine the fun with having to reach around and touch the emergency controls -- it sure won't be an 'alerter button' in a car that has to be steered as well as braked! -- every 40 seconds. Hope the driver or passengers weren't hoping for a restful journey!
erikem In addition to dealing with streetcars and RR crossings, I'd wonder if the the autopilot knows anything about "Turn around, don't drown" with respect to flooded roads.
In addition to dealing with streetcars and RR crossings, I'd wonder if the the autopilot knows anything about "Turn around, don't drown" with respect to flooded roads.
Train Nut AND Bike Nut Here.
Euclid Self-driving cars are dramatically over-promised. It seems like politically correct, utopianism. To jump right into this self-driving dream relying solely on computer technology is naïve. Here is what I don’t quite understand: Other than the fashion statement of have a car that drives you around automatically, what exactly is the point of idea? Is it to remove the fallibility of a human driver, and thus make driving safer? A few years ago, I wrote about my prediction for the future of driving where cars have a layer of automatic, central control that enforces the laws as you drive. I thought it would develop as a natural extension of the current proposals to tax drivers by the miles driven, which, in effect, turns all roads into toll roads. Instead of getting people out of their cars and into mass transit, it would turn people’s cars into a form of mass transit. This self-driving car development seems to have leapfrogged past my prediction and taken it to its ultimate level.
Self-driving cars are dramatically over-promised. It seems like politically correct, utopianism. To jump right into this self-driving dream relying solely on computer technology is naïve. Here is what I don’t quite understand: Other than the fashion statement of have a car that drives you around automatically, what exactly is the point of idea? Is it to remove the fallibility of a human driver, and thus make driving safer?
A few years ago, I wrote about my prediction for the future of driving where cars have a layer of automatic, central control that enforces the laws as you drive. I thought it would develop as a natural extension of the current proposals to tax drivers by the miles driven, which, in effect, turns all roads into toll roads. Instead of getting people out of their cars and into mass transit, it would turn people’s cars into a form of mass transit. This self-driving car development seems to have leapfrogged past my prediction and taken it to its ultimate level.
Of course running the entire road system under centralized control would deprive governments of a nice,steady revenue stream: speeding fines. Any toll-by-the-mile system would also be able to monitor average speed and automatically ticket speeders.
"I Often Dream of Trains"-From the Album of the Same Name by Robyn Hitchcock
Ulrich Completely self driving cars are still a few years off although cars that steer themselves down the highway are already here (check out the Mercedes class E).
Completely self driving cars are still a few years off although cars that steer themselves down the highway are already here (check out the Mercedes class E).
Given that vehicles with self steer also (in all cases AFAIK) have "smart" cruise control systems which automatically detect traffic and operate both the vehicles throttle and brakes I would say that for highway operation they qualify as "self driving". In the next few years there will be more advanced versions of the system that will automatically change lanes in response to traffic.
erikemA big factor was the design of the side stick control on the plane that did not give feedback as to what the other pilot was doing with controls. With the traditinal control yokes, the captain would have known immediately that the first officer was yanking back on the yoke and thus prolonging the stall.
Certainly part of it, and I am a much bigger fan of Boeing's approach to FBW and Otto. However, I think that the most important outcome of this crash was the understanding of the complex human factors that played a major role.
NorthWestWhat if the car suddenly gives back control when an icy skid becomes outside parameters?
erikemAll the more reason to keep the driver informed of what the car was doing, in particular feedback through the steering wheel. A co-worker suggested that the "auto-pilot" mode should do what's done on locomotives, have an alerter function that requires interaction from the driver at intervals to check if the driver is paying attention.
But then what problem are you trying to solve with automation? Unlike in aircraft there are far fewer efficiency gains besides increasing capacity of roads, which would probably reduce vehicle spacing to the point where a human reaction would probably be too slow. The premise for these systems has mostly been to avoid accidents caused by drunks, texters, etc, who are not paying attention to the road, and to allow more productive use of time in the vehicle by allowing the occupants to do other things instead of drive. If the driver has to constantly monitor the vehicle, then why not simply drive the thing instead of trying to deal with the issues of trying to pay attention, complacency and the startle factor resulting in improper reactions when the car suddenly hands control back?
My observations and experiences - either the 'machine' has total control or the human has total control. Shared control generally means that the human is in no position (mental and/or physical) to assume control when the machine decides to relinquish control. A human deciding to disengage machine control is one thing - a machine deciding to ceede control is a different animal entirely.
Never too old to have a happy childhood!
I have no doubt that self-driving cars can be technologically perfected. However, the concept seems to be about so much more than transportation. It seems to be a facet the green movement just like renewable energy. This virtue seems to be responsible it being pushed so hard by the public sector. It has that coercive feel of the banning of the incandescent light bulb.
Some of this is reflected in this piece:
http://www.usnews.com/news/business/articles/2016-01-14/government-developing-policies-for-self-driving-cars
Quotes from the link:
Foxx [of USDOT] said the government believes self-driving vehicles could eventually cut traffic deaths, decrease highway congestion and improve the environment. He encouraged automakers to come to the government with ideas about how to speed their development.
"In 2016, we are going to do everything we can to promote safe, smart and sustainable vehicles. We are bullish on automated vehicles," Foxx said during an appearance at the North American International Auto Show in Detroit.
Bryant Walker Smith, a law professor at the University of South Carolina and an expert on the legal issues surrounding self-driving cars, said the government's action is aggressive and ambitious. He said regulators are following the example of Europe, which has exempted autonomous cars from certain regulations in order to speed their development.
Foxx said President Barack Obama's budget would provide $4 billion over the next decade for programs to test connected vehicles. Some vehicles already on the road can communicate with each other, or with traffic lights or stop signs, through cellular signals. Eventually, that could prevent accidents, since vehicles could warn each other before they crash at an intersection.
It seems to me that the timeline for the perfection of the self-driving car is being incredibly over-promised in the name of the politically correct virtues of the green movement and its quest for sustainability. Most ironically, therefore, the driverless car is an assault on the automobile.
8 ways driverless cars will drastically improve our lives
http://www.techinsider.io/8-ways-driverless-cars-will-drastically-improve-our-lives-2015-12
One of the 8 ways quoted from the link:
In addition to saving lives, driverless cars may also help save our planet.
Because autonomous vehicles are built to optimize efficiency in acceleration, braking, and speed variation, they help increase fuel efficiency and reduce carbon emissions.
According to McKinsey, adoption of autonomous cars could help reduce car CO2 emissions by as much as 300 million tons per year. To put that into perspective, that’s the equivalent of half of the CO2 emissions from the commercial aviation sector.
Clearly the author of that article lives in the city.
Out here in the sticks, I don't see many of those touted advantages as being advantages. Even if you can call for a vehicle and have it arrive at your house, such functionality will rely on the sufficient availability of the vehicles. Will my decision to use one for a two hour trip to somewhere mean someone doesn't make it to work? Will one still be available ten hours from now when I want to return? What if someone at my destination wants to use one of the vehicles to travel even further from my home? Could one of these vehicles theoretically travel coast to coast, in bit and pieces by a number of individuals?
And if individuals don't own the vehicles, who does?
Larry Resident Microferroequinologist (at least at my house) Everyone goes home; Safety begins with you My Opinion. Standard Disclaimers Apply. No Expiration Date Come ride the rails with me! There's one thing about humility - the moment you think you've got it, you've lost it...
One way that i can see this working is on Private toll roads and HOV lanes- Imagine that we could repurpose abandoned railroad corridors into robot roads that have driverless electric trucks and buses serving industries and stations along the way-
https://techcrunch.com/2016/04/25/the-driverless-truck-is-coming-and-its-going-to-automate-millions-of-jobs/
Mercedes-Benz found out with one of their semi-self driving cars is that the lane detector doesn't work when the lines in the road are covered in snow. What I see happening eventually is that the Elon Musks of the world will give money to "help" politicians campaigns who will then enact laws where all federal, provincial and state roads must have readable-in-all-weather sensors installed in the middle and sides of roads and in crossings of various types, both railroad and pedestrian. This will be done in the name of safety and will be paid for by your increased taxes. Don't you like safety?
The idea of truck or car "platooning" has been shown to increase road capacity by reducing the space between cars.
http://arstechnica.com/cars/2016/04/europe-completes-a-demonstration-of-semi-autonomous-truck-platooning/
Back in the 1960's, GM experimented with a wire laid in the pavement. Their "self guided" cars had a sensor which detected the wire and followed it - rather like an electronic slot car...
A wristband that delivers a mild eletroshock when the driver nods off would be a start.
Our community is FREE to join. To participate you must either login or register for an account.