Millennium Line - Broadway Extension: Planning & Pre-Construction

[Reecemartin]

Originally Posted by trofirhen

By the M.Line built out, do you mean Arbutus and / or UBC? And is the "poco spur" not The Evergreen Line? just trying to clarify your information request

I mean to UBC and well the Evergreen Line was rolled into the M Line no? By the Poco spur I mean the spur that would run to Port Coquitlam from Coquital Central.

And to chime in on autonomous vehicles, I'd say we are probably in the middle. I'm doubtful that the tech will be 100% ready in two to three years but, in five I think that's pretty much guaranteed. It's true people have been talking about it for a long time but now we've achieved critical mass. All the big auto companies, Tesla, Uber, Lyft, Google, Delphi, Baidu and probably Apple are all working on autonomous cars now and competition in the market usually leads to faster progress. And when the market demands it I think regulations will be faster than we think.

[AMTDGT]

Originally Posted by Jebby

I hope to be proven wrong. I just don't see the CoV and BC regulators as very progressive.

Agreed With the power the taxi association has in the lower mainland, I doubt there will ever be Uber in Vancouver let alone fully autonomous cars. Hope I'm wrong

[cornholio]

Originally Posted by Reecemartin

I mean to UBC and well the Evergreen Line was rolled into the M Line no? By the Poco spur I mean the spur that would run to Port Coquitlam from Coquital Central.

And to chime in on autonomous vehicles, I'd say we are probably in the middle. I'm doubtful that the tech will be 100% ready in two to three years but, in five I think that's pretty much guaranteed. It's true people have been talking about it for a long time but now we've achieved critical mass. All the big auto companies, Tesla, Uber, Lyft, Google, Delphi, Baidu and probably Apple are all working on autonomous cars now and competition in the market usually leads to faster progress. And when the market demands it I think regulations will be faster than we think.

You mentioned all the little auto companies but most of the traditional big ones have been incorporating autonomous features for a while. For example VW etc. Even Skoda (under the VW umbrella) has had autonomous features for a couple years now.
Here is a budget Skoda from 2 years ago.
• Video Link

[Reecemartin]

"....mass. All the big auto companies, Tesla....."
Lol

[VancouverOfTheFuture]

autonomous driving has been around for decades. people just forget. there are 2 huge hurdles to get this thing mainstream.

1 - governments. they will want years and hundreds of thousands of test miles, maybe even over a million, before they legislate it to go full blown. and we saw with the guy who died driving the Tesla, the system can make mistakes. (big white truck was seen by the car as a cloud).

2 - cost. these drive systems are expensive, and they aren't nearly comprehensive enough to fully drive on their own. the need for human input is still very high.

with those 2 in mind, i would say 10 years is a hopeful estimate. people forget how governments don't like being liable for things, and if anything were to go wrong due to technology being rushed, and governments allowed it anyways; that would not end well.

The Eureka PROMETHEUS project was then the largest research and development program in history when it comes to autonomous driving vehicles and related technologies, with a total cost estimate of EUR 749 million in current money.

Mercedes-Benz made a number of driverless prototypes that culminated with a re-engineered W140 S-Class that technically drove almost entirely by itself over 1,678 kilometers (1,043 miles) from Munich to Copenhagen back in 1995.

Autonomous Driving Technology

[Kisai]

Originally Posted by VancouverOfTheFuture

autonomous driving has been around for decades.

There's been steady improvements, most of these will be rolled into what we formerly referred to as "cruise control"

Like if you've watched any TV ads for cars lately, you'll note that a great deal of attention is put on safety things like back-up cameras, lane-assist, auto-park, and so forth.

The Skytrain is really the thing that proved that you can have a completely automated train, no drivers. Applying the exact same concepts to road cars would solve this quickly.

While Google and companies work on what I'll just refer to as "assisted autopilot", what really needs to come to the table is a dedicated real-time CBTC-like moving block system for cars. These can either be mandated by insurance (cars without autopilot simply plug in a box to their CANBUS and that puts them on "the network" so auto-pilot cars and traffic control systems know where they are. Cars with built-in systems simply tell the driver when they've left the network and to be ready to take over if needed.) The next step is making sure that cars that "fall off the network" or are from outside the network or too old can still be driven.

Like that is one way to ensure that LRT/street cars don't need drivers and work with the existing traffic systems. You make sure that every car licensed to drive in those cities is on the traffic control network.

But I digress, we're likely to see the automated cars as an answer to the last-mile problem of rapid transit before the price of owning one comes down.

[Reecemartin]

Originally Posted by Kisai

There's been steady improvements, most of these will be rolled into what we formerly referred to as "cruise control"

Like if you've watched any TV ads for cars lately, you'll note that a great deal of attention is put on safety things like back-up cameras, lane-assist, auto-park, and so forth.

The Skytrain is really the thing that proved that you can have a completely automated train, no drivers. Applying the exact same concepts to road cars would solve this quickly.

While Google and companies work on what I'll just refer to as "assisted autopilot", what really needs to come to the table is a dedicated real-time CBTC-like moving block system for cars. These can either be mandated by insurance (cars without autopilot simply plug in a box to their CANBUS and that puts them on "the network" so auto-pilot cars and traffic control systems know where they are. Cars with built-in systems simply tell the driver when they've left the network and to be ready to take over if needed.) The next step is making sure that cars that "fall off the network" or are from outside the network or too old can still be driven.

Like that is one way to ensure that LRT/street cars don't need drivers and work with the existing traffic systems. You make sure that every car licensed to drive in those cities is on the traffic control network.

But I digress, we're likely to see the automated cars as an answer to the last-mile problem of rapid transit before the price of owning one comes down.

You make an excellent point, the type of utopian self driving car future where we see cars driving bumper to bumper at 140 kmh on all highways and where the cars are all aware of other cars, is going to require significant collaboration. Self driving cars on their own are cool but, when you link them all together in a network that's when you create a sea change.

[cganuelas1995]

Originally Posted by Reecemartin

You make an excellent point, the type of utopian self driving car future where we see cars driving bumper to bumper at 140 kmh on all highways and where the cars are all aware of other cars, is going to require significant collaboration. Self driving cars on their own are cool but, when you link them all together in a network that's when you create a sea change.

I hope the security on them is flawless because I can see hackers/terrorists/youtube pranksters having wet dreams of this.

[Reecemartin]

Originally Posted by cganuelas1995

I hope the security on them is flawless because I can see hackers/terrorists/youtube pranksters having wet dreams of this.

Flawless no, in IT nothing is flawless, however it doesn't need to be flawless. It only needs to be good enough to be safer than the current system. Ideally though yeah security needs to be tight.

[swimmer_spe]

So, about that Broadway Extension...

[Reecemartin]

Originally Posted by swimmer_spe

So, about that Broadway Extension...

Haven't heard much as of late in the news...

Will be pretty cool when some of these projects start getting underway, as we will see quite a few projects being completed in succession.

[aberdeen5698]

Originally Posted by Kisai

The Skytrain is really the thing that proved that you can have a completely automated train, no drivers. Applying the exact same concepts to road cars would solve this quickly.

While Google and companies work on what I'll just refer to as "assisted autopilot", what really needs to come to the table is a dedicated real-time CBTC-like moving block system for cars.

Wow. Oh wow! What a woefully oversimplified view of the problem you have!

In your world we could easily solve the self-driving problem by developing an inter-car communication protocol and fitting every car with a transponder to broadcast its location. The rest is just algorithms.

The problem is that autonomous cars have so, so much more to deal with than merely other, well-behaved autonomous cars!!! They have to worry about pedestrians, they have to worry about cyclists, they have to worry about not being able to see the roads or painted lines due to rain and snow, they have to worry about broken-down vehicles blocking the roads, they have to worry about road construction.

The reason Skytrain is so easy is because we've spent good money building a separated guideway which removes all those problems.

The biggest challenge of autonomous driving isn't the software that tells the vehicle how to drive. It's the software and sensors that can reliably tell the vehicle where and what and the obstacles are.

[Reecemartin]

Originally Posted by aberdeen5698

Wow. Oh wow! What a woefully oversimplified view of the problem you have!

In your world we could easily solve the self-driving problem by developing an inter-car communication protocol and fitting every car with a transponder to broadcast its location. The rest is just algorithms.

The problem is that autonomous cars have so, so much more to deal with than merely other, well-behaved autonomous cars!!! They have to worry about pedestrians, they have to worry about cyclists, they have to worry about not being able to see the roads or painted lines due to rain and snow, they have to worry about broken-down vehicles blocking the roads, they have to worry about road construction.

The reason Skytrain is so easy is because we've spent good money building a separated guideway which removes all those problems.

The biggest challenge of autonomous driving isn't the software that tells the vehicle how to drive. It's the software and sensors that can reliably tell the vehicle where and what and the obstacles are.

Seems a little rude Imo, "in your world" anyways I think he did indeed over simplify but, he brought up a good point. All the obstacle avoidance etc, is being worked on by pretty much every company working with autonomous vehicles. The point is networking them together is no simple task and will require alot of collaboration, and from what I know there isn't a standard for this being developed right now.

[aberdeen5698]

Originally Posted by Reecemartin

The point is networking them together is no simple task and will require alot of collaboration, and from what I know there isn't a standard for this being developed right now.

And my point is that it can't be a networking problem because cars will have to operate in an environment where many of the obstacles aren't going to participate in a network. If cars have to operate with even some non-networked obstacles, then the issue of networking is moot. Cars need to detect obstacles on their own, and since that's the case networking isn't going to solve any real problems.

It's much, much more difficult than just a lack of a network.

[Reecemartin]

Originally Posted by aberdeen5698

And my point is that it can't be a networking problem because cars will have to operate in an environment where many of the obstacles aren't going to participate in a network. If cars have to operate with even some non-networked obstacles, then the issue of networking is moot. Cars need to detect obstacles on their own, and since that's the case networking isn't going to solve any real problems.

It's much, much more difficult than just a lack of a network.

The point of a network has little to do with actual obstacle avoidance, that's something individual cars can handle, but saying networking isn't going to solve problems isn't true because imagine you have an obstacle obstruct a highway lane. Networked cars in that lane could know that they need to slow down at the same time that the car immediately risking collision would, I can think of lots of places where cars having knowledge of other cars not immediately surrounding them could be useful.

Anyways to bring us back to the Millenium Line, since this is the first ART subway in Vancouver where we get to build the infrastructure completely ground up, do we expect stations to be built to a much higher spec than the Canada Line? IMO I'd like to see 2 points of vertical movement of the platform. Also do we expect island or side platforms?

[aberdeen5698]

Originally Posted by Reecemartin

Anyways to bring us back to the Millenium Line, since this is the first ART subway in Vancouver where we get to build the infrastructure completely ground up, do we expect stations to be built to a much higher spec than the Canada Line? IMO I'd like to see 2 points of vertical movement of the platform. Also do we expect island or side platforms?

Oh, I so hope so. There should be at least two access points on diagonally opposite corners to every station, IMHO. If I had my way then there'd be four, one on every corner, the way that Toronto does it.

I'm not really expecting that level of construction, but I hope that the fact that the Canada Line has significantly exceeded expected passenger loads right from day one will help to convince the people holding the purse strings that it's in our best long-term interests not to under-build this line.

[Reecemartin]

Originally Posted by aberdeen5698

Oh, I so hope so. There should be at least two access points on diagonally opposite corners to every station, IMHO. If I had my way then there'd be four, one on every corner, the way that Toronto does it.

I'm not really expecting that level of construction, but I hope that the fact that the Canada Line has significantly exceeded expected passenger loads right from day one will help to convince the people holding the purse strings that it's in our best long-term interests not to under-build this line.

Also I never have understood much why we could have a four corner system for the Canada Line, if you look at NYC or Toronto they usually have one "main" entrance to the mezzanine, and then each corner has a somewhat narrow staircase that exits onto the sidewalk like you mentioned. Surely is it's just a small staircase like this it can't add THAT much to the cost. I see Broadway as having significant future growth and so it would be wise (common sense tbh) to not shoot ourselves in the foot with this...

Interestingly Toronto has the opposite problem, over on the UrbanToronto forums people regularly criticise the TYSSE-Toronto York Spadina Subway Extension (http://www.ttc.ca/Spadina/index.jsp rightly so....) For building maddeningly large white elephant style stations in largely undeveloped areas. While I don't want the TYSSE, some things here are actually not at all bad ideas, the stations are spacious, offer multiple entrance and exit points, and all have public art, even with some very cool stuff hanginign above the platforms.

[aberdeen5698]

Originally Posted by Reecemartin

...Toronto has the opposite problem, over on the UrbanToronto forums people regularly criticise the TYSSE-Toronto York Spadina Subway Extension (http://www.ttc.ca/Spadina/index.jsp rightly so....) For building maddeningly large white elephant style stations in largely undeveloped areas. While I don't want the TYSSE, some things here are actually not at all bad ideas, the stations are spacious, offer multiple entrance and exit points, and all have public art, even with some very cool stuff hanginign above the platforms.

Toronto understands that if you're going to go to the expense of building an underground line it makes sense to spend a bit of extra money during construction to provide large stations and ample access to them. Because trying retrofit it later when it's needed (and it will eventually be needed) would be far, far more expensive.

That's my big beef with the Canada Line. It's as if they never expected it to be very popular, so why spend money on the ability to handle lots of people? It's small town thinking, and Vancouver deserves better.

[Reecemartin]

Originally Posted by aberdeen5698

Toronto understands that if you're going to go to the expense of building an underground line it makes sense to spend a bit of extra money during construction to provide large stations and ample access to them. Because trying retrofit it later when it's needed (and it will eventually be needed) would be far, far more expensive.

That's my big beef with the Canada Line. It's as if they never expected it to be very popular, so why spend money on the ability to handle lots of people? It's small town thinking, and Vancouver deserves better.

Spot on

The importance of Central Broadway cannot be understated, and it's growth is only going to speed up with the construction of a Subway, the current system downtown might be able to get by on a lesser spec because, it's mainly for commuting because all the stations are so close together, but for Broadway we really need to make sure the stations have a ton of capacity. Just look at the Canada Line, half the problem there isn't even just the short trains but, the TINY stations that totally hamper flow and thus capacity. Hopefully we don't see another post Evergreen Skytrain Line that doesn't offer up AND down escalators (in part this is why island platforms are good, we can have more centralized space for up down movement). In addition hopefully these stations are planned for higher traffic in their fare gates as well, iirc Bridgeport has like four fare gates. As a comparison Wellesley on the TTC (a relatively minor station whose passenger numbers are not much more than Bridgeport) has something like 10 fare gates.

TLDR:
People need to go to community meetings etc. and push for the M-Line extension to be built to a much more significant spec than the Canada Line this line could vary well see many more riders, and everything from platforms, to escalators, to entrances, to fare gates need to be planned as such.

[Kisai]

Originally Posted by aberdeen5698

Toronto understands that if you're going to go to the expense of building an underground line it makes sense to spend a bit of extra money during construction to provide large stations and ample access to them. Because trying retrofit it later when it's needed (and it will eventually be needed) would be far, far more expensive.

That's my big beef with the Canada Line. It's as if they never expected it to be very popular, so why spend money on the ability to handle lots of people? It's small town thinking, and Vancouver deserves better.

But Toronto's Subway primarily connects their universities to downtown universities and the Via/Amtrak rail, it doesn't connect their Airport.

I'm pretty sure the entire purpose of the Canada Line was to be nothing more than an Airport Express for the 2010 Winter Olympics, and it was the only thing that could be built in time. Had Richmond been on board a decade ago it might have been planned better and started years earlier.