Autonomous will never happen, and absolutely never happen in the UK.

Do you not think there is a reason why mm wave 5G is being introduced (other than to spread covid, of course)?

Some people have some strange ideas about 5G, I wonder what people will say when 6G comes in?

There have been some who have been concerned about 5G (as well as 2G 3G 4G Wi-fi). This is long before COVID-19 happened. Unfortunately, the issue has been hijacked by conspiracy theorists.

There was a Senate hearing and Snr Blemthal asked the industry if 5G was safe. Big wireless concedes no studies showing safety of 5G (7th Feb 2019).

www.youtube.com/watch?v=vZ5soLrvXFg

I doubt a science researcher is going to getting funding to investigate 5G. Even if they found concerns..... they would be putting their reputation on the line.

I turn off my wi-fi at night (you can log in). You keep your router on, but turn off the wi-fi.

If you have the latest 5G enable phone, you can select to use non-5G networks.... (not tried it).

5G just gives faster download speeds.....

Fully autonomous is inevitable - the only question is when.

Fear of change is partly about loss of control and fear of the unknown. Other irrationalites. Sail out of sight of land and fall off the edge of the earth. Trains at high speed will cause the body to melt and render cows infertile. Cars need to be preceded by a man with a red flag.

Autonomous does not mean 100% safe - merely better than fallible human beings who will continue to suffer from anger, inattention, alcohol, emotions,tiredness etc etc.

Software improves in function and reliability over time. With the huge amounts of data available for every collision a process of continuous improvement will can eliminate weaknesses.

The UK will not be the first place that autonomous becomes commonplace - that is likely to happen where (a) road networks are less idiosyncratic, (b) traffic levels are lower, (c) legal constraints more readily overcome.

In this country my guess would be that autonomous units will be introduced first in controlled environments - eg: airport car park to terminal, then on to motorways where there are no pedestrians and all vehicles move in the same direction. I would put initial automation at 3-7 years, and full automation at ~10 years.

It isn`t inevitable by a long way, processors are capable now of controlling a car, but the amount of data needed is far more than we have at the moment, and growing, so will be many years before it it does become commonplace assuming companies don`t give up on it due to costs, also software constantly has to be updated

Fully autonomous is inevitable - the only question is when.

Fear of change is partly about loss of control and fear of the unknown. Other irrationalites. Sail out of sight of land and fall off the edge of the earth. Trains at high speed will cause the body to melt and render cows infertile. Cars need to be preceded by a man with a red flag.

Autonomous does not mean 100% safe - merely better than fallible human beings who will continue to suffer from anger, inattention, alcohol, emotions,tiredness etc etc.

Software improves in function and reliability over time. With the huge amounts of data available for every collision a process of continuous improvement will can eliminate weaknesses.

The UK will not be the first place that autonomous becomes commonplace - that is likely to happen where (a) road networks are less idiosyncratic, (b) traffic levels are lower, (c) legal constraints more readily overcome.

In this country my guess would be that autonomous units will be introduced first in controlled environments - eg: airport car park to terminal, then on to motorways where there are no pedestrians and all vehicles move in the same direction. I would put initial automation at 3-7 years, and full automation at ~10 years.

The problem Terry is that the boffins at motor companies have been saying much the same for the last 20 years now. Scientists have said the same about AI, nuclear fusion, space missions to Mars and a whole host of other things for decades now, and yet...

Computerised systems are normally very good at doing a few things in a controlled environment. What they aren't good at is rapidly assesing and learning how to operate (whatever) in a 'new' environment or one that rapidly changes, particularly where we animals and nature in general interract with said (whatever), often in what appears to be unusal, almost ramdom ways.

Until computer systems can actually learn as humans can do across a wide range of skill sets and environments and rapidly adapt (AIs currently cannot), then we will still be in charge of cars.

My estimate, if fully autonomous cars ever come about (i.e. AI happens) is at least 30 years, more like 50 and more. I suspect it won't help that the current circumstances we face will likely bankrupt the world, meaning there will be little incentive to produce autonomous cars for just a few thousand uber-rich people.

I wonder what is happening with driving in other industries - say driving boats. I fancy going on a sailboat trip but don't want the hassle of managing sailings myself. So AI driven navigation and sailing control would be nice.

Just remember how much time and space is required to stop or alter the course of a supertanker ?

Science fiction writers say brain will become larger as humans have to work hard to control AIs but limbs will shrink small because of not much use.

The idea of Ai is so we don`t have to think, it learns as it goes and takes the place of your brain because you won`t need it, like robots might take over everything we do which is gradualy happening

The vulnerability already exists to hackers, terrorists etc etc. Perhaps we should go back to the 1960's (pre-electronics era) or even the horse and cart

Maybe they will start using the system windows 11 will, so everything installed on car computers will be encrypted so hackers can`t get into it.

it might to some be a pain in that all software and hardware replaced on a car will need a cryptographic key to install on the car and without it it won`t install.....

Many new cars are already connected via sat-nav, phones, location, operating data to garages, tracking devices etc.

Current software already provides lane departure warnings, emergency braking, cruise control, road sgn recognition. All quite safety critical.

Functionality is maintained and updated via internet or (sometimes) a computer and cable.

The vulnerability already exists to hackers, terrorists etc etc. Perhaps we should go back to the 1960's (pre-electronics era) or even the horse and cart (although the feed may be tampered with!).

Nothing wrong with the 2000s cars:

People not getting their car nicked by someone with a scanner in a car park/hiding in your front garden etc for you to use the keyless entry system;

Cars not needing endless 'software updates' that provde little benefit or that are done because the car wasn't ready to go into production and owners were used as beta testers;

Cars not beeping at you every few seconds because it notices something's 'there' or carries out an emergency stop in the fast lane of the motorway because a plastic bag has got caught in the front grille/bumper area;

Not having to go through ten menus to switch radio stations etc;

Not having to remortgage my home if I wanted to replace a lightbulb on the car (not that I've had to pay for that yet) or electronic gizmo I never use;

Very reliable car.

No chance is some hacker being able to 'take control' via a phone, etc etc

Having it shod with sensible, but still decent (for driving/ride quality) wheels and tyres that last a long time and don't cost a fortune to replace (because they are very common).

It cost me £10k to buy new after discounts. A similar car today (15 years later, forgetting the pandemic price increases) would likely cost me best part of £20k, and that's an ICE car.

Still drives almost as well today as it did when I bought it in early 2006. Costs me peanuts to run.

No inbuilt satnav etc (perfectly fine ICE), but can use my newest smartphone just clipped onto the air vent as a decent satnav when required.

iThe hackers are still keeping many steps ahead of the software developers and security experts, with no sign that will change in the future.

If the hackers are a step ahead we'd not be able to use any IT at all as it would already be compromised

HOW DO YOU KNOW IT ISN'T

My PC had a Russian Mafia virus for quite a while before (after gradually getting suspicious) I put a sniffer on it. None of the AV suites I tried would detect or disinfect (Clamwin found it once) but I THINK I killed it by reforrmatting. They don't necessarily advertise.

Unless auto pilot achieves a maturity where it is possible to sleep while car is driving of itself, it is more of trouble than anything else.

It is lot more difficult to let the car drive itself but then observe and be alert in case you have to take over - compared to - you just drive yourself.

Autonomous driving as an upgrade of cruise control (i.e. adaptive) is fine but truly driverless car for all types of roads is still fairy tale.

Yep. Been saying this for ages now. So-called 'AI' computers/programs are very good at doing specific, often very limited-in-scope tasks, but not very good at multi-tasking and especially adapting to new/unusual circumstances and instantly.

It's the reason why 'driverless' trains are far easier to design, but even then they often require 'staff' on routes other than airport shuttles in case a system breaks down/an accident so a real human can take over/make decisions as required to deal with the situation and the passengers.

Moreover, if you're relegated to having to catch up on sleep in your autonomously driven car, perhaps think about your life choices.

Unless auto pilot achieves a maturity where it is possible to sleep while car is driving of itself, it is more of trouble than anything else.

Report in today's Times mentions 17 collisions of Teslas in US in the last 3 years, mostly with emergency vehicles at night. Perhaps fast-moving flashing lights confuse the system ?