Source: BBC News
Google is to start building its own self-driving cars, rather than modifying vehicles built by other manufacturers. The car will have a stop-go button but no controls, steering wheel or pedals.
This seems like a pretty exciting move, even if the vehicles are limited to 25mph. A self-driving car with no manual operation at all sounds like a great improvement and I'd definitely be interested in buying, renting or riding in one (especially if there were fleets of self-driving taxis available at short-notice in London).
Google have announced that they're now going to be tackling tricky busy city streets rather than the somewhat more straightforward roads they've been operating on till now. This is important for cities like London (or, in fact, almost any city in Europe) where the roads are often narrow and clogged with both moving and stationery vehicles.
One area in particular that I'm keen to hear about is how the system will cope with pedestrians. Not sensible pedestrians who use pavements and pelican crossings, or even drunk pedestrians who stagger along the pavement and occasionally collapse in the gutter. I'm not worried about children darting without thought into the road or slow-moving pedestrians limping slowly across when the cars have right of way. All of these things can be anticipated, managed, handled and once the crisis is over everything returns to normal and the car goes about its business again.
The type of pedestrian which, to me at least, presents the most interesting use case is the chap who, recognising a self-driving car with excellent brakes and computer-fast reactions, simply steps into the road trusting the car to stop in time to miss him. Unlikely, you think? It already happens in London, to a certain extent, on roads that are busy with pedestrians like Long Acre where taxi speeds are already constrained and it's reasonably easy to force the drivers to slow further or stop. Think how much worse it will be when the pedestrians know (even if this "knowledge" isn't, in fact, correct), rather than merely believe, that the car will stop.
The taxis eventually make it down Long Acre and along Wellington Street but it can take a very long time and they're often only able to do it by nudging across pedestrian crossing or through gaps that, strictly speaking, they shouldn't attempt. Will a self-driving car be able to break or bend the rules to make progress through heavy pedestrian traffic? It seems most unlikely.
By my reckoning there will have to be radical change. Roads that are currently open to traffic at all hours will have to be restricted if we are to avoid total gridlock. Zebra crossings, which work brilliantly for pedestrians, will have to be re-thought in areas where there are enough pedestrians to stop traffic completely. Some vehicles, like coaches, HGVs and construction vehicles, will have to be limited by time or location and it's not going to be pretty.
In short, if you're expecting self-driving cars to solve transport problems in inner-city areas, think again. And get ready for some bigger changes.
The type of pedestrian which, to me at least, presents the most interesting use case is the chap who, recognising a self-driving car with excellent brakes and computer-fast reactions, simply steps into the road trusting the car to stop in time to miss him. Unlikely, you think? It already happens in London, to a certain extent, on roads that are busy with pedestrians like Long Acre where taxi speeds are already constrained and it's reasonably easy to force the drivers to slow further or stop. Think how much worse it will be when the pedestrians know (even if this "knowledge" isn't, in fact, correct), rather than merely believe, that the car will stop.
The taxis eventually make it down Long Acre and along Wellington Street but it can take a very long time and they're often only able to do it by nudging across pedestrian crossing or through gaps that, strictly speaking, they shouldn't attempt. Will a self-driving car be able to break or bend the rules to make progress through heavy pedestrian traffic? It seems most unlikely.
By my reckoning there will have to be radical change. Roads that are currently open to traffic at all hours will have to be restricted if we are to avoid total gridlock. Zebra crossings, which work brilliantly for pedestrians, will have to be re-thought in areas where there are enough pedestrians to stop traffic completely. Some vehicles, like coaches, HGVs and construction vehicles, will have to be limited by time or location and it's not going to be pretty.
In short, if you're expecting self-driving cars to solve transport problems in inner-city areas, think again. And get ready for some bigger changes.