Self Driving Car thread

[M. Jones]

Alamout[/url]":37dooe0f]You can pretty much do that already, with mass transit instead of self-driving cars. In fact you can do it much better: no matter how good self-driving cars get, they won't be able to match the passenger density of a train.

..the passenger density of a full train.

Trains are a tradeoff between capacity, overhead, frequency and manpower. For the time being one of the bigger advantages of trains is that enclosed routes can be fully automated, as long as you can prevent passengers from falling on the tracks or getting stuck in closing doors.

 

[M. Jones]

Alamout[/url]":30q6ipmh]That process would take many decades to happen organically--longer than self-driving cars will exist, perhaps. Unless we develop revolutionary advances in both the cost and time of construction projects. Societies and governments need to cause this amount of change--markets aren't going to do it on their own, not at a useful timescale.

This misperception explains your biases, at least.

I can't think of too many times when governments reacted to market needs faster than markets did. Free societies are markets. Here you're talking about taking advantage of technology to increase efficiency and decrease costs. Normally the action of governments is to retard efficiency for political reasons -- jobs, incumbent financial interests, populist appeal.

 

[M. Jones]

Pont[/url]":3g82pj0u]
I really think self-driving taxis will be the biggest, most profound change brought on by self-driving car technology.

They're basically PRT that utilises existing infrastructure, and as such were predicted in these forums to be the eschaton of mass transit.

 

[M. Jones]

redleader[/url]":kj4yujen]
No, there are essentially no crosstalk issues with LIDAR. However, lets be clear, its not even certain that lidar will be the technology of choice. 60GHz microwave is probably cheaper and equally as good as optical because you can use electronic beamforming whereas lidar must be mechanically scanned or have multiple fixed emitters/detectors. If you want cars very close to each other, a single 360 degree phased array (or 2 180 arrays fore and aft) are much faster than a galvo scanned optical beam and certainly cheaper than many lidar beams.

Radar must normally be licensed, with the FCC. 60GHz is unlicensed, but mostly so because it's severely attenuated by resonance with atmospheric oxygen, and thus of severely limited range. Rain fade is a problem with all EHF spectrum, according to Wikipedia.