Autonomous cars

Suppose your car could drive itself. The first benefit would be that you could read or work or talk on the phone ( safely ) while you were carried to your destination. Another is that computer controlled cars would move more efficiently on our highways – allowing more people to move at a higher speed over the same roads – reducing commute times. But one of the most interesting is that you wouldn’t need to keep it with you all the time.

Your car could come when you called it. In the morning you drive to work, your car can go elsewhere – it can go home and serve someone else in your family. A lot of families instead of needing 2 cars, would only need 1.

Another huge benefit is that we could eliminate most parking lots. Huge areas of office/retail space wouldn’t need parking lots. That would allow greater density and then all the trips for everyone are shorter. Imagine the area that you work or shop without parking lots – everything is closer together and many things are walkable that weren’t before.
Cars would still need to park while they waited to be called, but those parking lots can be 5-10 minutes away from places people want to be – and they can be packed nose to tail – valet style.

But beyond that – there is an even greater benefit. You wouldn’t even need to own a car at all. You could pay per car minute, with a premium for service guarantee. You could pay a lot for a car guaranteed within 5 minutes of your call – but someone willing to plan in advance or use cars at off peak times would pay a lot less.

My commute is about 20 minutes each way when traffic is good. I use 40 car minutes per day. Assuming 12 core dailylight hours – thats about 6% of a car per day.
If you assume that the entire workforce commutes in a 3 hour window in the morning and a 3 hour window in the evening thats about 15% of a car per day. ( Also assuming that the car only takes a few ( 7 ) minutes to get to its next user )
If you assume a 2 hour commute window and 30 minute commute+return time, then its 25%.

The number of cars necessary would then be between 1/4 and 1/6 of the number in use now to move the same number of people during the rush hours.
If the autonomous cars can get you to your destination faster than human driven, then that will reduce commute times, and reduce the number of cars needed.

Another opportunity for efficiency is that one person trips can be made by one person cars. You can pack 2-4 one person cars into the space on the highway that one general purpose car would take. You call for the type of car that you need for each trip, if you have a party of 8 – you could get an 8 person vehicle ( or two fours ). If no 1 person car is available, you get the next size up.
The fleet of cars could be made up lots of different vehicles to meet the peak/average needs.
Here are some total guesses blending the needs of the commute peak vs other general uses:
30% 1 person vehicles
20% 2 person vehicles
30% 4 person vehicles
15% 6 person vehicles
5% 8 person vehicles
You get the flexibility of being able to use an 8 passenger vehicle when you need one, but since most of your trips are 1 or 2 people, you use the more efficient vehicle at that time.

The energy used by a one person vehicle to take you on your trip and then return to its holding area or to its next user will be dramatically lower than the energy used to move a larger general purpose vehicle one way and then park.

Another huge efficiency gain is available because the system can usually know your travel distance ahead of time and provide a car with the appropriate range.
If you are going to go 10 miles from home to work in a 1 person vehicle, you can be provided with a 1 person car with a 10 mile range. If you want to go 300 miles, you get a 300 mile range vehicle.
If you assume that the 1 person car is going to be very busy for a 3 hour period each commute period – it will provide 6 person trips, probably a total of 120 miles ( including distance traveled empty between pickups ) then you want to have vehicles with a 120 mile range so they can serve throughout the entire commute period.
At that point you could trade off having one 120 mile range car in the system or multiple shorter range cars. The two 60 mile range cars would use less energy, but at the expense of another vehicle. There might be a sweet spot having 120x 100 mile range cars instead of 100x 120 mile range cars.


2 responses to “Autonomous cars

  1. Car-sharing of this sort would work particularly well in regions like ours where we have a two-way commute. More people definitely live on the west side and commute east, but there are plenty of offices downtown. A car could pick someone up in Bellevue, drop them off in Pioneer Square, pick me up on Queen Anne, and then drop me off in Bellevue.

    This sort of car-sharing also opens up the possibility of a carpooling service to many more businesses. Right now a business has to be Google or Microsoft sized to have its own vanpool (or in Microsoft’s case, full bus service.) If you could skip the drivers and not have to own the cars even a company of a couple hundred people would be able to offer reasonably timed carpools for its people who live near each other. There are at least ten of us living in about a two mile radius on the west side that could use a van to get to work if that service were available.

  2. A couple of other points of efficiency:

    • Vastly reduced accident rate. An automated system is probably not going to be trouble-free, but it will do way better than the average human being does.

    • Much more efficient use of surface streets. A while back, I saw a simulation of traffic control that didn’t use any sort of traffic signaling, but rather simply allocated intersection usage according to car size, travel lane, and speed. It was harrowing to consider being in a car passing through an intersection designed with that in mind, as each car going through would miss the others by the merest of clearance, rather than each direction of travel being batched up for passage through the intersection. But traffic flowed _much_ more efficiently.

    I could swear I read about a pilot project somewhere in which this idea of automated, shared transportation units was being explored.

    Unfortunately, I cannot find links to either of these references at the moment. But I’m sure they exist. 🙂

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s