I wrote the initial interface specification for the co-ordination system in Rio's open-pit mines. The mere fact that we were able to account for the state of every moving thing in the pit makes this a completely different proposition to self-driving vehicles in an uncontrolled real-world environment. At a guess it's at least ten to a hundred times easier.

Eventually this could be put into place in the broader traffic ecosystem - if all the cars on the road were required to have a transponder sort of arrangement on board, you could have a central point of control, and perhaps tie in people's phones etc to locate pedestrians. I do want to note that I am aware that there is an unfathomably large amount of complexity behind that however.

There's a movement towards this; look up "Vehicle-to-Everything" or "V2X". The idea is to enable things like one vehicle transmitting sensory information or intent to others around it, helping to prevent rear-end collisions, etc. My personal take is that it's well intentioned, but there will always be things that aren't part of the corpus of communicating devices that will then limit the utility. I suspect the communication of sensor data is the more useful part than some of the demos I've seen so far of, for example, a car sending an alert to a person's phone so they don't step into a driveway when it's reversing.

Because my dog will totally have his mobile phone on him when I'm chasing him down. Or my little kid chasing after his ball...

And when software on a car assumes pedestrians always have a mobile device on their person, charged and ready for a split second alert? IMO This would be an absurd design.

It'd be a tool in the toolkit, in my hypothetical. I mean, by the time autonomous cars were common enough on the road, we might all be tagged with an rfid chip as a normal identification or something... I don't disagree that it's an absurd idea, but it's just a devils advocate position I'm discussing here.

One can also see elevators as self-driving electric vertical motion vehicles. Sure, they are not operating on public roads, and the extra constraints reduce complexity, but they've been around for a while.

Personally, I don't see mining operations as anything remotely close to public roads autonomous vehicles because you have a controlled environment in which paid workers that have been trained to interact with equipment and be aware of its limitations and dangers are performing the said interactions in a strictly constrained manner with the autonomous equipment clearly visible as such and while protected by all kinds of insurance and work-specific regulations.

There's none of that even in the testing stage of current "autonomous" driving vehicles.

Very large work sites are far from completely controlled environments. Visitors, wildlife, weather, equipment malfunctions, human error, and a host of edge cases from trespassers to protesters.

See neighbor comment from someone who actually knows of what they speak: https://news.ycombinator.com/item?id=17184043

Visitors, trespassers and protesters are all a very remote possibility in the Rio Tinto mines mentioned, and certainly wouldn't be a surprise if they did show up.

No real traffic and predetermined path cleared of unexpected obstacles.

All vehicles can safely communicate with each other and are controlled from central location.

It is automated driving but not even close to "self-driving".

> Self driving vehicles have been used at Rio Tinto's Pilbara mine for a couple of years.

Accidents still happen up there.

https://www.youtube.com/watch?v=5WZCweaakOg

(*video may not explicitly of an autonomous truck, I'm not entirely sure, but just using it as an illustration)