Self-Driving Cars

[Nick2]

So I thought I'd bring up this just to talk about. For the most part, everone on this forum would agree that public transit is the best thing for smart growth and city development. However, for a city set up like Charlotte or any other city that isn't as dense as New York or Chicago, cars will always be king. Rail transit is massively expensive. Perhaps that money could be better suited to improving railways. Now before you get your pitchforks and torches ready, I'm looking towards the future.

Google has made huge strides with their automated cars in the last couple of years. We could see them implemented on a major scale realistically in the next two decades. That means more cars, more efficiently, more quickly. For somewhere as dense as south end and uptown, rail transit as well as good pedestrian access is crucial but in somewhere like university city, it might be smarter to find ways to improve the road grid. 

Smart cars might never come to fruition but if it's the way of the future, it'll look silly investing in a technology from two centuries ago when it was made obsolete in two decades after being built.

This topic is just to start a discussion about this.

[Matthew.Brendan]

The real advance with self driving cars will come once we start building sensors and other tech into the roadways themselves. Right now these self driving cars are reactionary and driving on "dumb" roads. 

[tarwater]

I've been wanting to make this topic for a while. I think you're right that in a city like Charlotte, in 2016, investing any more money in long-term rail transit plans doesn't make a bit of sense. Too little, too late. We won't need it, and almost no one will want to use it even given the option. People like to get where they are going quickly, and without interacting with strangers whenever possible. A fleet of self-driving cars can move people around faster, and take them right to the doorstep of their destination. Imagine a very cheap, ubiquitous Uber service with no driver. It's not far off. 

Personal car ownership is also going the way of the dinosaurs, and parking lots/decks along with it. Good riddance. What are we gonna do with all the extra space? 

1 hour ago, Matthew.Brendan said:

The real advance with self driving cars will come once we start building sensors and other tech into the roadways themselves. Right now these self driving cars are reactionary and driving on "dumb" roads. 

The REAL advance is when we get the dangerous, reactionary, human drivers out of the road altogether.  

[kermit]

Meh, their are several basic physical and economic laws that indicate that self-driving cars can't be the total game changers that their fans project them to be:

• Newton's laws tell us that individual metal boxes will always be energy inefficient.
• Cartesian geometry reminds us that there is not enough two dimensional space to accommodate individuals heading to the same employment district at one time (and business and workers have clearly demonstrated a preference for clustered employment rather than dispersed employment).
• Neoclassical economics tells us that self-driving cars will be more expensive than transit (see Newton's laws) -- marginally higher costs for a daily commute might be OK for the affluent, but will be a deal breaker for the middle and lower classes.
• Economics also tells us that eliminating the fixed-cost element of owning a car is likely to make people people think harder before spending $5 on a ride in a driverless urber for short trips when walking will be a cheaper alternative -- I honestly think self-driving cars will increase demand for walkable spaces since no one will have a car sitting in their driveway which gives the impression of nearly zero marginal cost for short-trips.
• Urban housing will also get cheaper since no one will need to spend $10,000 and up to pay for their bundled parking space.

My personal perspective is that the only good part of driving is "driving" so sitting in a small metal box that I have no control over has zero appeal to me.

[southslider]

Land is limited. Self-driving transit makes more sense. 

[cjd5050]

On 8/19/2016 at 4:43 PM, Matthew.Brendan said:

The real advance with self driving cars will come once we start building sensors and other tech into the roadways themselves. Right now these self driving cars are reactionary and driving on "dumb" roads. 

Personally I think it goes beyond this.  While automated driving has a relational connection to the road and drivers around it, it's still a temporary cache of information.  Where the big change is going to come from is 'big data' and pattern recognition.  

While humans are 'dumb' they are also pattern driven.  We have the simple patterns like someone goes to work M-F at the same time and on the same path but that's just where it begins.  Once you starting adding peripheral changes like daylight and irregular changes like weather or special events those patterns change.  Then if you consider how connections are made and understand that something happening in the transportation network in San Diego, Houston or Orlando will eventually impact Charlotte...it really starts to become an amazing puzzle.  Humans will adapt to these changes of course but will do so  in their own way but when doing so all coordination is lost.  

An example of patterns is 3 walkers, including myself, in my neighborhood.  I walk my dog every single morning at 6am.  I need to because that is when she is used to going to the bathroom.  My first neighbor down the street is a senior and she walks every morning but only when the sun is out.  So her schedule changes as daylight changes.  My second neighbor is a work from home dad.  He will walk his baby every morning at the same time provided the weather allows.  These are 3 individual but predictable patterns.  Right now we have the ability to analyze this type of data and output changes but we don't have the ability to capture the data.  

Roll this up to vehicles, not just cars but busses and trucks, and you can start to see the potential.  A container ship docking in Charleston can be scheduled but where the containers go is lost data right now.  Add in you can't control the weather and you can start to see the problem.  The containers still have to go from A to B but when they leave, what path they take could all be eventually scheduled and automated.  This is because once you capture enough data you can find predicability in the patterns and then coordinate the unpredictable.  This would mean a truck that needs to go from Charleston to Charlotte might be able to leave the port at 9am but actually it would be a better result if it left at 9:20 because of how it would interact with the entire network along the way.  In turn it would either arrive in less time or the trip would be more fuel efficient.  All the while doing this on an existing road network that is not even close to being at full capacity if you remove dumb humans from the equation.   

 

 

 

 

[SgtCampsalot]

Just one more very useful tool in the toolbox for improving our human habitats. Not a deus ex machina, but like you say, neither are trains. It all needs to be done in conjunction.

[ah59396]

Found this pretty applicable for what we are talking about here:

 

http://www.curbed.com/2016/8/31/12691516/self-driving-bus-vehicles-finland-helsinki-transportation

[cjd5050]

1 hour ago, ah59396 said:

Found this pretty applicable for what we are talking about here:

http://www.curbed.com/2016/8/31/12691516/self-driving-bus-vehicles-finland-helsinki-transportation

Those look awesome.   How could would it be if each light rail station had 10 of these that went out different directions for a mile or two?  Essentially running the same loop all day long.  Would really expand the accessibility to light rail.

[ah59396]

22 minutes ago, cjd5050 said:

Those look awesome.   How could would it be if each light rail station had 10 of these that went out different directions for a mile or two?  Essentially running the same loop all day long.  Would really expand the accessibility to light rail.

I think we are getting closer and closer to that reality.  Hopefully we live in a future where mass transit is complimented by autonomous vehicles.

[kermit]

A very sober assessment of self-driving tech at wsj (not paywalled)

http://www.wsj.com/articles/self-driving-hype-doesnt-reflect-reality-1474821801

[videtur quam contuor]

When I see those "self driving" cars my first thought is that 50,000$ worth of high tech equipment sitting right there on the roof, waiting for someone to nick it at any opportunity.

[tarwater]

If you think Uptown has too much wasted space now, wait until every parking deck/lot is sitting empty because no one drives any more. This is 10 years away.

Article from Forbes on this issue: https://www.forbes.com/sites/alanohnsman/2018/05/18/end-of-parking-lot-autonomous-cars/#6c97dab07244

[CM2424]

On 5/18/2018 at 3:51 PM, tarwater said:

If you think Uptown has too much wasted space now, wait until every parking deck/lot is sitting empty because no one drives any more. This is 10 years away.

Article from Forbes on this issue: https://www.forbes.com/sites/alanohnsman/2018/05/18/end-of-parking-lot-autonomous-cars/#6c97dab07244

If the parking lots and decks went away the land would be used for something else that can generate revenue - housing, office space, retail, etc.  This is a massive positive change for Charlotte and all other cities