South Korean Capital Launches Self-Driving Bus Experiment 29
South Korea's capital launched its first self-driving bus route on Friday, part of an experiment which engineers said aims to make people feel more comfortable with driverless vehicles on the roads. From a report: The new vehicle does not look like a regular bus and has rounded edges along with large windows that make it appear more like a toy than a technological breakthrough. This design is intentional, said Jeong Seong-gyun, head of autonomous driving at 42dot, the start-up responsible for the self-driving technology that is now owned by auto giant Hyundai. "This is the future," he told AFP, adding that the bus required "a considerable new type of design." The bus looks a bit "like Lego" and is made of composite parts to help keep costs down and make it easy to replicate, he said. It uses cameras and lasers to navigate the way instead of expensive sensors, Seong-gyun added. The company's goal was to make the technology low-cost, safe and easily transferable to many types of vehicle in the future, for example delivery trucks.
Half the size, twice the frequency (Score:4, Interesting)
Once you eliminate the driver, the economics of running the buses changes.
You can make them half the size and run twice the frequency, and/or run additional routes.
Re: (Score:3)
And in fact this thing is not a full-sized bus, so the premise checks out.
Buses are terrible. The only reason we use them is that drivers are expensive. When that goes away, so do the full-sized buses.
Re: (Score:2)
Buses have 40 seats but on average carry 7 passengers.
Re: Half the size, twice the frequency (Score:2)
Sure, but what's the peak?
Re: (Score:2)
Sure, but what's the peak?
Small buses can handle the peak periods by running more often since we can afford more of them.
But they save energy and road space during off-peak hours.
The net effect is fewer empty seats and more convenience for passengers.
2*(1/2 x) > x (Score:3)
The upside is that more "half buses" means the bus comes more frequently, dramatically reducing your wait time (at the stop or inside because the bus isn't scheduled to arrive for x minutes).
Re:2*(1/2 x) x (Score:4, Interesting)
Two "half buses" still cost more than one bus with respect to both capital and maintenance costs.
False. A quality transit bus is around $500,000+, a quality midibus (two of which carry as many people as a full size bus) is $250,000 or less. The difference is even more pronounced at the low end; a cheapie midibus is about $80,000-100,000, while a cheap bus is $250,000+. The maintenance costs will be higher for two buses, but the initial purchase price is actually lower.
In addition, I would presume the capital costs of a self driving bus are higher than a human-driven bus of the same size, but am not certain of that assertion.
They certainly are, because you need the self-driving equipment. But it's very expensive to pay for a human driver over the lifespan of a bus, it actually costs more than the bus in most cases. So if the vehicle delivers on its promise, the self-driving equipment can be quite expensive and still offer lower TCO as compared to a human-driven bus. As well, the maintenance costs of buses should be substantially lower for EVs, and the fuel costs are definitely much lower. Buses are actually designed to be maintained, unlike cars, and there is usually great access, so doing suspension upkeep is likely cheaper per mile traveled. You need specialty tools to do that kind of stuff on a light vehicle these days. Just try changing the ball joints on a Sprinter without either a special holding fixture for your press, or the official forcing tool. Hence the value of a minibus or midibus over a cutaway shuttle, which is based on a light vehicle...
The upside is that more "half buses" means the bus comes more frequently, dramatically reducing your wait time (at the stop or inside because the bus isn't scheduled to arrive for x minutes).
Yes, and there is another upside, being able to run buses at higher capacity (meaning better return on your capital costs) at off hours when there are fewer riders... by reducing per-vehicle capacity. And then there's the massive reduction in road damage...
bus driver also acts as an kind of security guard (Score:2)
bus driver also acts as an kind of security guard as well.
With no driver in some areas bad things can happen on the bus with no one to enforce any rules.
Re: (Score:2)
A security guard who rides the bus is cheaper than a bus driver, so if you feel you need one of those, the TCO might still be lower.
Re: 2*(1/2 x) x (Score:1)
...the self-driving equipment can be quite expensive and still offer lower TCO as compared to a human-driven bus.
You're not factoring in the cost of gutting-out said technology, either to replace it with newer less flawed tech or simply to put a driver's seat, steering wheel and pedals back in where they belong.
That these are foregone conclusions is self-evident to all but shills, fools and accountants - i.e. fools.
Re: (Score:2)
Grand proclamations are peachy keen, and they sound excellent, but who knows what's going to happen? Maybe where drivers are respectful and roads are clean, self-driving on predefined routes will work great. This certainly isn't the first autonomous bus project, it's not even the first one this year [singularityhub.com]. And then there's this [cavforth.com], I like to imagine it spoken in a really bad, exaggerated Scottish accent. "For many years people have been waiting with bated breath for the promise of completely autonomous vehicles" ...
Re:2*(1/2 x) x (Score:2)
Two "half buses" still cost more than one bus
A quick Google search shows wholesale prices of $60k for a 40-seater and $37k for a 20-seater. .
I would presume the capital costs of a self-driving bus are higher than a human-driven bus
The SDB doesn't need a driver's seat, steering column, mirrors, or even windshield wipers.
The only additional equipment are cameras and a computer. The cameras are very cheap. Lidar is expensive, but Teslas operate without Lidar.
Re: (Score:2)
Lidar is expensive, but Teslas operate without Lidar.
Teslas run away and run over motorcycles without LIDAR because they are shit at estimating depth, actually worse than a human in fact. The cost of LIDAR is negligible compared to the cost of a bus (a good school bus is over $200k, even for one with a medium duty diesel, and transit buses cost more) and also compared to the cost of just one accident.
Teslas don't use LIDAR because Musk is a dipshit, not because it makes sense.
Re: (Score:2)
The bus can't be self driving 100% of the time.
Maybe someday it can, but for now people are already operating these vehicles by telepresence. Also, it's easy to imagine a sort of semi-automatic mode which you could control with a gamepad, or even a phone (although even a bad gamepad would be better than that.) You could drive at low speed only with it, until the vehicle cleared the obstacle that it couldn't handle.
Re:2*(1/2 x) x (Score:2)
I would presume the capital costs of a self driving bus are higher than a human-driven bus of the same size
The capital costs of the vehicles are dwarfed by the cost of paying the drivers.
Over the lifetime of the vehicle the drivers will be paid many times the cost of the vehicle; therefore you can purchase (and operate) several (even if higher capital cost) driverless vehicles for less than the cost of purchasing one lower cost vehicle and paying the drivers.
Re: (Score:2)
A bus driver may do more than drive (Score:2)
Who is going to check the tickets, enforce the rules etc.
Re: (Score:3)
Who is going to check the tickets, enforce the rules etc.
This wouldn't work everywhere, but it will probably work fine in Seoul, or Tokyo, or a number of other places. One person can remotely monitor multiple buses for signs of trouble, and dispatch police or maintenance crews as necessary. Also, there's lately a push towards fare-free public transportation in many areas, and if you reduce costs then it becomes more feasible to operate in this manner. Eliminating fares increases ridership, which decreases traffic, emissions, etc. It's cheaper and more feasible th
Re: (Score:2)
in other city's remotely monitor multiple buses = some one who reviews the tapes after something has happened for the cops.
Re: (Score:2)
Semi-self-driving vehicles can already be remotely monitored and controlled [mashable.com], this is going to be necessary for the foreseeable future in order to handle unusual cases.
Oblig... (Score:2)
There are already running Level 4 Buses (Score:1)
Infrastructure (Score:2)
In the late 90s, MIT did research on self driving cars. They found that there was a very simple way of implementing fully autonomous driving. All you had to do was drive a ferromagnetic nail into the roadway every foot. This let any vehicle simply and easily align itself, path-find, and navigate. The cars themselves just used IR passive motion detectors and ultrasonic sonar to detect anything in the roadway; both cheap and well established technologies even in the late 90s. The computational load was light.
Re: (Score:2)
Re: (Score:2)
What's next? Self flying planes? (Score:2)
Ow wait...