Autonomous Buses and Other Autonomous Vehicles

[BusHunter]

I was watching an assisted parking video on youtube and discovered in a few google search clicks that the US is slated to have it's first autonomous bus soon in a office park in California.

http://gizmodo.com/the-uss-first-autonomous-buses-will-drive-around-a-cali-1734989938

In fact there's a few cities around the world that already have this on an even bigger scale. Probably the most impressive is the Yutong bus, which actually goes up to 40 mph and travels real streets in traffic. It has a driver behind the wheel just in case it fails but he works at observing the system do it's thing only.

http://gizmodo.com/5-cities-with-driverless-public-buses-on-the-streets-ri-1736146699

In fact this technology seems closer than we think. Tesla Motors has went as far as having an update to it's software in it's cars that enable existing cars already purchased to go semi autonomous already. They don't want to say it's autonomous buts it's hands free and operates sort of like an auto pilot. It will change lanes on it's own and keep speed, although some of the luxury cars and even some almost entry level cars already have this option. I know Ford and Chrysler do. To go fully autonomous, approval is needed from the government, but they claim at Tesla this could be by 2020.

http://www.dezeen.com/2016/01/11/elon-musk-predicts-completely-autonomous-driverless-tesla-cars-in-two-years/

Uber claims they want to have autonomous taxi's by 2030 at even cheaper prices for the consumer and google wants to get in on the autonomous taxi market possibly sooner. Detroit automakers claim the technology could come as soon as 2025 in it's cars, and by 2040, 75 percent of all cars will be autonomous. This technology could really change the transportation landscape whether it's buses, cabs or personal vehicles. One article even said that they predict more cars with be car sharing cars and possibly sales of cars in general will go down. It's an interesting topic to discuss to say the least!!

[MTRSP1900-CTA3200]

7 hours ago, BusHunter said:

...In fact this technology seems closer than we think. Tesla Motors has went as far as having an update to it's software in it's cars that enable existing cars already purchased to go semi autonomous already. They don't want to say it's autonomous buts it's hands free and operates sort of like an auto pilot. It will change lanes on it's own and keep speed, although some of the luxury cars and even some almost entry level cars already have this option. I know Ford and Chrysler do...

Glad you mentioned Tesla. I saw a video of two people playing patty cake while their Tesla cruised down a a highway. Cars nowadays have the hardware. I was looking at the new Ford F-150. That thing has cameras all over the place. Somehow they even managed to make a 360 degree camera.

Of course you can also read about the adorable Google autonomous car.

[Busjack]

You had previously discussed the Freightliner autonomous truck here. Car and Driver had an article about semi-autonomous cars, noting that those other than Tesla had a tendency to veer into the merge lane.

Most of the buses in the articles you cited operate in a dedicated lane. Aside from whether an autonomous bus could operate in traffic, the real question is whether it could recognize potential passengers at the bus stop, or could only go "station to station," like the Loop Link.

[BusHunter]

That could be remedied with a beacon type system like Pace was trying out. If the beacon were tripped the bus would know to stop there. If you notice the riders on the small bus which is electric too are pressing a button in the video to open the doors. The only problem I see is how would the bus be able to enforce someone to pay a fare. If you notice the test is where the buses are free. Maybe this could work in that respect unless they hired conductors and shed operators, but they'll most likely figure out something. The bus could always just disable if it didn't get a fare. If buses were eventually with no drivers I wonder then if frequencies would increase or more all night service would exist. They wouldn't be paying for someone to drive an empty bus. The fares could even fall. I think we may actually see this hit the North American TA's by another 10 years at least on the light routes.

Going back to the point of autonomous vehicles on the road being legal. (each state has to legalize this before it can be done) Nevada and Arizona seem to be among the first states that are trying to legalize driverless vehicles, so this is where we'll probably see them first. 

[Busjack]

8 hours ago, BusHunter said:

. If you notice the riders on the small bus which is electric too are pressing a button in the video to open the doors. The only problem I see is how would the bus be able to enforce someone to pay a fare.

That is theoretically easier to answer for the first rider--you need a valid Vemtra card to open the door. Not quite as easy to answer if the second rider can sneak in while the (bing bong) doors are closing.

Maybe a bus becomes only semiautonomous for the same reason as the Freightliner truck--to make sure that the driver doesn't plow into 205 N. Michigan Plaza.

[BusHunter]

2 hours ago, Busjack said:

That is theoretically easier to answer for the first rider--you need a valid Vemtra card to open the door. Not quite as easy to answer if the second rider can sneak in while the (bing bong) doors are closing.

Maybe a bus becomes only semiautonomous for the same reason as the Freightliner truck--to make sure that the driver doesn't plow into 205 N. Michigan Plaza.

They could put something like a rotogate (or revolving door if a rotogate is too ugly) on the bus. This way if the rider doesn't pay he don't get on either and riders leaving would be able to leave without him getting on. Doing it this way you wouldn't need a ventra card reader outside the bus but inside at the fare entrance. The bus wouldn't leave the stop until it reads the rotogate is cleared or no one paid a fare or activated the doors. Instead of a button at the doors why not just put a supermarket type style electronic eye door opener. Rig each stop to send out a signal to the bus that this is indeed a stop so it wouldn't just open the doors on the fly. Lots of possibilities here. This proves to be the next big thing for technology on the bus after the electrics come out. 

[Busjack]

7 minutes ago, BusHunter said:

They could put something like a rotogate (or revolving door if a rotogate is too ugly) on the bus.

I was thinking something like a revolving door in front of a store, but a similar idea. However, how long would it take to complete boarding? Maybe you have come up with another reason why these are used mostly on shuttles. The China video only shows that the driver can operate "hands off" and not rear end a car, but doesn't show passengers boarding or paying fares.

[BusHunter]

8 minutes ago, Busjack said:

I was thinking something like a revolving door in front of a store, but a similar idea. However, how long would it take to complete boarding? Maybe you have come up with another reason why these are used mostly on shuttles. The China video only shows that the driver can operate "hands off" and not rear end a car, but doesn't show passengers boarding or paying fares.

Then you have the wheelchair riders but surprisingly the small bus can deploy a ramp, but the revolving door would have to be wide enough for wheelchairs, maybe a system of doors is easier or just put the wheelchair position up by the operators position. Put the wheelchair position between the two doors for ease of boarding. Of course the bus wouldn't move until it reads a wheelchair is fastened in. The bus still wouldn't move if the bus reads people between the first and second door. Only wheelchairs would be excluded. 

[Busjack]

8 minutes ago, BusHunter said:

Then you have the wheelchair riders but surprisingly the small bus can deploy a ramp, but the revolving door would have to be wide enough for wheelchairs, maybe a system of doors is easier or just put the wheelchair position up by the operators position. Put the wheelchair position between the two doors for ease of boarding. Of course the bus wouldn't move until it reads a wheelchair is fastened in. The bus still wouldn't move if the bus reads people between the first and second door. Only wheelchairs would be excluded. 

I hadn't thought of that. But then it gets to the question whether the wheelchair passenger needs the driver's assistance in becoming secured and putting on the seat belt.

[Busjack]

This discussion reminds me that the one thing that is not on a Pace Bus Operator job description is "drive a bus."

[Pace831]

47 minutes ago, Busjack said:

This discussion reminds me that the one thing that is not on a Pace Bus Operator job description is "drive a bus."

That's sort of implied by "transport passengers". But it does still say "dispense and collect transfers", even though they don't have to do that anymore.

On the issue of how to make sure people pay the fares, the system of doors seems like it could slow down boarding enough to significantly increase travel time which could even become a disincentive to using the system. How much space would the system of doors take up, and would that reduce the number of seats available? Would there need to be controls to manually drive the bus if necessary, and if so how much more space would be required? Would there be concerns about increased crime without an operator present? What to do if someone won't pay the fare and, consequently, the bus won't move? And surely the unions will take issue with the bus drivers' jobs going away. The only scenario I can see this working is on a BRT with off-board fare collection, otherwise there are too many "what if" possibilities.

[Busjack]

14 minutes ago, Pace831 said:

That's sort of implied by "transport passengers"

Maybe, but in the context of this discussion, it isn't, which is why I brought it up. One can toss the dice whether HR foresaw this trend at the time of drafting it.

The one serious point which led me to this tangent was my prior one on assisting the wheelchair passenger, which points out all the other things an operator has to do that have nothing to do with controlling the vehicle.

[BusHunter]

The bus could call the police or transit security officers and there could be police call boxes on board sort of like the #5000 call boxes.

They could make the system completely off board paying, then no doors are needed problem solved. They probably would have to make a glass shelter box maybe either with the screen windows for ventilation or a wall with holes in it. (like downstairs at Cermak/Green line south entrance) or a self ventilated/heated system. Put the rotogate on the shelter and the opening doors on the streetside that can only open with a bus present. Then that may work.

It probably wouldn't cost so much to do this, probably the buses would be expensive anyway and be delegated to only a few routes. It could be a slower infrastructure build, but still the Ventra cash machines would have to remain at the train stations. I think they are a theft risk unmanned.

It will be interesting anyway to see how the current BRT stations will become off board paying. Alot can be learned by the engineering of those stations.

[Busjack]

7 minutes ago, BusHunter said:

They could make the system completely off board paying, then no doors are needed problem solved.

 

8 minutes ago, BusHunter said:

It probably wouldn't cost so much to do this, probably the buses would be expensive anyway and be delegated to only a few routes.

 

8 minutes ago, BusHunter said:

It will be interesting anyway to see how the current BRT stations will become off board paying. Alot can be learned by the engineering of those stations.

But that gets back to my original point that maybe this only works for station to station operation, not flag stops at bus stop signs.

[BusHunter]

27 minutes ago, Busjack said:

 

 

But that gets back to my original point that maybe this only works for station to station operation, not flag stops at bus stop signs.

I don't see how it wouldn't work, especially on lines like #88 or #55A. They could actually get double their fun doing #55N and #55A because half the routes run together with the same stops. 

They wouldn't have operators so scheduling could be tailored to the beacons activated on route. The money would be there, not paying operators would save them tons of money.

Are you saying it would add alot of extra time to the run?  If fares are prepaid, maybe not.

[Busjack]

3 minutes ago, BusHunter said:

I don't see how it wouldn't work, especially on lines like #88 or #55A. They could actually get double their fun doing #55N and #55A because half the routes run together with the same stops. 

They wouldn't have operators so scheduling could be tailored to the beacons activated on route. The money would be there, not paying operators would save them tons of money.

Are you saying it would add alot of extra time to the run?  If fares are prepaid, maybe not.

No, what I am saying is that if you were converting these routes to autonomous bus, you seem to be saying that the prepaid boarding areas would have to be built first.

Aside from the cost of building various structures on Higgins or Archer Heights, there might be the problem, which I raised in connection with the original plan for the Jeffery BRT (not Jump) of a prepaid area in a cage with a rotogate and gate over the curbside that the prepaid passengers would be sitting ducks for a drive by shooting.

[BusHunter]

I can't control a drive by, but riders would have the option to not go into the shelter until they saw the bus was coming if they had anxieties about that or claustrophobic reactions to being closed in. The shelters would probably have to be bigger anyway because how would you let rear door riders exit? But if the bus is small like an optima it doesn't really need a rear door. They might need the space anyway. 

As far as not building shelters until afterwards what are you saying, riders will ride for free for a period of time? Nothing would be built anyway unless it was confirmed your TA was getting the bus. Usually it takes a year to get a bus, they could build this in a year on small routes. If they were desperate they could cut some stops until they got those online if construction was too slow or just let the customer ride for free. (bad business proposition though!! )

[Busjack]

This is becoming an interesting brainstorming session. Bringing up "all that construction," it might have been cost justified for a route like Jeffery with an articulated bus every three minutes, but not really on these fringe routes.

[Pace831]

Google's self-driving car hits a bus.

[artthouwill]

Even a computer can't anticipate what another driver will or won't do. 

[Busjack]

15 hours ago, artthouwill said:

Even a computer can't anticipate what another driver will or won't do. 

True, but there is enough radar (and inter car telecommunications are foreseen) that it will stop the car a heck of a lot quicker than you can.

[BusHunter]

1 hour ago, Busjack said:

True, but there is enough radar (and inter car telecommunications are foreseen) that it will stop the car a heck of a lot quicker than you can.

It also comes down to can you trust a car to do the right thing and if it doesn't are you still liable for the accident? Now it might be alright if Google or whoever pays for your accident repair, but no one wants a damaged car either even if someone else is paying for it to get fixed. Nothing is better than the original parts, body. Me personally, I feel that if the car is going to be in an accident, I want to be behind the wheel because as demonstrated here if complex scenarios arise, will the autonomous driver be able to troubleshoot it's way out of an accident? I've had some really close calls driving, but a good driver will make the right decisions and avoid an accident. An extra set of eyes also can help avoid an accident. I look at these autonomous features right now as a novelty and wouldn't use them unless I had to. You can still park a car faster than a computer and sometimes the computer will fail as demonstrated here.

https://www.youtube.com/watch?v=J5d1YIc7kE4

I wonder if a tight park is needed will it attempt to park or tell you it can't do it.

[Busjack]

24 minutes ago, BusHunter said:

I wonder if a tight park is needed will it attempt to park or tell you it can't do it.

Supposedly, the radar in the car can measure the parking space.

Most of this stuff, like the rear view camera, are there because you can't see out the back window due to the hiked up trunk or you are in a SUV way above the next car's hood. M-B demonstrated cameras for this purpose at the Auto Show about 20 years ago.

[BusHunter]

33 minutes ago, Busjack said:

Supposedly, the radar in the car can measure the parking space.

Most of this stuff, like the rear view camera, are there because you can't see out the back window due to the hiked up trunk or you are in a SUV way above the next car's hood. M-B demonstrated cameras for this purpose at the Auto Show about 20 years ago.

The bad part of automation is that it makes you a worse driver than you were making you lazy and not able to troubleshoot effectively. Airliners have got so advanced a plane can land itself, but if it's autopilot fails and the crew is incompetent like what happened on Air France 447 the plane just goes down.

This makes for tons of discussion, like do we let people slide on parking cars at the DMV because a computer now does it? What may slowly happen is you get less and less people qualified to be on the road that only pass half of the rules of the road or can do it on a written test but cannot do it real life.

[Busjack]

4 minutes ago, BusHunter said:

This makes for tons of discussion, like do we let people slide on parking cars at the DMV because a computer now does it?

At some point you use an automated shared ride device and don't need a drivers' license. The question basically goes to the interregnum.