Probably the first stage to be reached will be autonomous driving on specific, designated roads - motorways, dual carriageways. The real challenges are urban roads.

However, taking a journey on a motorway where the software draws not only information from the immediate environment (adjusting distance from the vehicle in front, lane changing and so on), but is fed with the larger picture about traffic density up to several miles ahead, could make for a more relaxed drive and, if enough vehicles are using the system (because at first there will be a mixture of driver-controlled cars and autonomous ones) the journey could actually be quicker and more fuel-efficient.

Think most people know they won't be here that quickly - just the car companies that are developing that think it will be quick which is ually the way with companies. Fully autonomous cars are a huge challenge but will come eventually.

It really comes down to how big company pockets are in R&D and if there is a chance the public will accept them, I`m now sceptical they will take off and you are more likely to see autonomous robots than cars, they will be more usefull and labour costs will be nill using them?

The development of autonomous cars is occurring at the same time as the development of electric cars, and the two are somewhat interlinked because electric car technology lends itself to autonomous technology.

. . . But I don't see the two trends continuing together at the same rate.

The move to electric cars has begun in earnest and I can only see the pace picking up on this.

But while there are autonomous cars running in trials around the world, they are not foolproof yet and there's not the same public desire for autonomous cars as there seems to be for electric ones.

Instead the autonomous technology will continue to spin off into driver aids, intelligent speed limiting and suchlike, some way short of true 'driverless'.

However, it might be that the only manufacturers that can make safe autonomous cars are Japanese.

This could be interesting then.....

media.jaguar.com/news/2018/03/waymo-and-jaguar-lan...g

I think you missed the point I was making. It’s all very well Jaguar making an announcement, but we know how unreliable cars are, including Jaguar. Do you really think they can be reliable enough that self driving will work? I don’t.

>>All the examples of fully automated trains I can think of are on self-contained railways designed from the outset for automation<<

Exactly - how many of our roads are designed for automated machines ? How many of the users on those roads have been trained for their use ? The industry appears to have chosen one of the most complex automation processes possible, and wants to do it before any other large scale automation involving untrained users. Let's do easier projects first and try to learn something in the process.

Thanks for the details on the Xiaomi robot vacuum - I'm pleased to see some projects actually produce something usable.

I don't think energy source (electric or oil) is connected to the method of control - they are independant but coincidendentally being developed at the same time

I do expect however that autonomous will be fitted to electic vehicles first if for no other reason than govts around the world are legislating againg ICE.

And whilst as individuals we may consider ourselves top notch drivers - alert, aware, thoughtful etc - the realitty is that most are not - hence 3000 fatalities on UK road pa.