Saw some in B&M yesterday........ £14.99.
Google something called Cooperative ITS. ITS stands for Intelligent Transport Systems. The 'Cooperative' bit refers to cars that interact with the road infrastructure, exchanging data on where they are, their intentions, road conditions and so on. Effectively, every vehicle on the road becomes a probe/source of data, not just a recipient.
Years away? Nah. I was at a conference in Detroit last autumn where GM's head said she'd accelerate deployment onto some roads within a couple of years. The same's going on in Europe.
As to camera systems, you might start by looking at CMOS sensors - the types of cameras you see in mobile phones. A CMOS-based camera the size of a 2 Euro coin could be stuck in a traffic signal post and provide a usable street-monitoring capability. That can easily be integrated into a vehicle and the quality of images your current smart phones can generate demonstrates a base level of image quality that we can expect. The cost point on the vision technology is coming down all the time. The car manufacturers are chasing a cost point in the single hundreds of dollars but it will continue to come down. Moore's Law and all that.
Cost is an issue, certainly. Look at things like the Terramax - the autonomous logistics vehicle that the US Army has been developing - pricey, but iterative generations and volume production will bring that down. That's the absolute top-end - a vehicle that can navigate off-road, rather than just following the road.
But cars such as Mercedes's S-Class already have machine vision technology on board which will allow the car to follow, unaided, the lines in the road, the cars in front and so on. The machine vision technology can pick out of the vista in front of the vehicle such things as road signs, interpret them and act/react accordingly. So the technology's already here for hands-off driving and at a price point that can go into top-level cars. Expect the E-Class to have it in the not-too-distant, and equivalent trickle-down in other manufacturers' ranges soon.
The sticking points are legislative - the Vienna Convention stipulated for a long time that there be a human in the loop. That's been changed recently, albeit on the quiet, but in reality cars such as the S-Class can already effectively drive themselves.
Google might be grabbing the headlines because it's a consumer brand but the mainstream motor manufacturers have been plugging away, individually and collectively through government-funded ITS projects, for years.
So, expect it be around a lot sooner than some of you expect.
As you say interconnectivety already exists..........as does the ability to alter what communications are made, what individual cars think they are/should be doing and various ways to give them wrong info or steal them.
Ignore the scare mongering?
With respect, a lot of effort has gone into making things robust. Too much is at stake if things are compromised. Ignore the scare-mongering.
I've been involved in this area for 15 years. A lot of the reason why this stuff has taken and is taking so long to come to market is the (absolutely correct) attitude of 'test, test, test and test again'.
There is some phenomenally badly informed writing even in the mainstream media. Some 'informed' writers think that any vehicle with an OBD2 port is instantly vulnerable. That's errant nonsense.
Yes, it's right to be concerned. No, we shouldn't be complacent. Yes, there will be attempts at hacking. No, it hasn't happened under anything other than controlled conditions. Yes, the developers have seen and learned and countered. No, breaking in will not be a pushover - far from it.
Sorry, I didnt see this post before my above reply.
Breaking in won't be a pushover...........yes ok, OK, sweet dreams
That's why its on the increase!
We're at crossed purposes here. I'm talking about hacking and taking control of an autonomous vehicle on the move, not theft.
You did mention breaking in with you OBD2 ref, but no, we arent at cross purposes here. On the move is possible too, whether high tech or low tech.
If you work in the industry you should know how as well!
Lots of things are sold as safe, and the public normally believe it, its just a shame that crooks decide to use equal amounts of skill to defeat the safe systems.
Lets go back to the OBD2 issue for a second. How many ARRSErs here do you think have a car with an immobiliser and a category one alarm system and so think a thief would need a programmed key to start and drive away their car?
You will know a thief wouldnt need a proper key to steal that car and that a screwdriver and £4 clone chip will do the job nicely.
Going back to moving vehicles that RECEIVE INFO from surrounding vehicles or traffic systems I can see room for meddling there
Crooks will adapt to the new systems like they do with every other technology. Who would have thought a decade ago that you could have your ID stolen by a crook using your fridge or kettle ect to gain access to your details.
Crash for cash may become stop for car theft at the low tech end. There are already ways to disable or confuse moving cars via wifi.
Its rather sad/amusing to realise that governments with assets like the CIA, MOD, NASA ICBMs, health services, national banks etc can be hacked or interupted by serious hackers yet the public believe they can buy a £18.000 car and it will be safe
On the other hand, there are and will be low tech solutions to defeat crooks.
If I hade a new roller or top end Merc (not likely lol) I'd be keeping the keys in a miniature faraday cage while in the house as thats the only current way to keep them safe.
Gone are the days where a thief has to actually enter your house and steal those keys to start and steal your car. They are just as usuable from within the house.