The prerequisite technology for an automated bus system is not the ability to take the driver out of the bus. The prerequisite is the ability to credibly signal to all the potential passengers what buses are going where they want to go, where they should get on those buses, and where they are supposed to get off.
All of the automation of bus routes involved is simple computation and I don't see AI having anything fundamentally new to add to the solution. You don't need to actually solve traveling salesman or something to add value to the bus system.
I say "credibly" because you really want the system to work almost perfectly right out of the gate. Otherwise passengers will shun it because the very baseline requirement of a transport system is to reliably get people where they wanted to go.
That would work pretty well. Just hail a bus with your phone, it tells you where to walk and what bus to get on, and where to get off on that bus, and how to finish your trip on foot or by connecting to another bus. It could all easily be integrated into Apple or google maps. No more fixed bus routes, just a bunch of AI buses driving around filling needs dynamically.
We can’t staff our fixed routes today, our agency is flush with cash but they can’t hire enough bus drivers with that cash, so they are cutting routes anyways forcing more more people back into cars. So obviously they couldn’t handle wide spread smaller buses with dynamic routing…maybe if they take a big subsidy on para transit, but that’s niche.
All of the automation of bus routes involved is simple computation and I don't see AI having anything fundamentally new to add to the solution. You don't need to actually solve traveling salesman or something to add value to the bus system.
I say "credibly" because you really want the system to work almost perfectly right out of the gate. Otherwise passengers will shun it because the very baseline requirement of a transport system is to reliably get people where they wanted to go.