Hacker News Reader

Comment by sheeshkebab | Hacker News Reader

>"If you assume the cost of these autonomous vehicles, the very early ones, will be six figures, there aren’t very many retail customers that are willing to go out and spend that kind of money," Ableson said.

Oh well I guess it will be a few years before I can recline and take a nap in my robo car while in traffic

View on HN

You'll never own a robo car, Uber and Lyft and Tesla and GM will.

Why not? It's a technology problem and technology almost magically gets cheaper over time. I don't see why it won't eventually be like all the other required safety features (airbags, ABS, etc).

If you live in the middle of SF it might sound silly to own an autonomous vehicle, but if you live out in rural Sonoma County it is going to sound a lot less silly. I frequently go out to the mountains and I doubt I'm going to want to call an Uber to load all my camping gear into and be gone for a week.

I live in the suburbs and I go on road trips all the time with the family.

About half of my neighbors own RVs.

Self-driving vehicles would be a dream come true.

Furthermore, if it's your own vehicle, fully autonomous mode under certain pre-defined conditions and roads can still be a huge win. But that constraint makes the same vehicles basically uninteresting as a shared vehicle taxi service for the near term.

It's great that all this work is being done but IMO summoning Johnny cab to take you home from a bar in SF is decades away.

Every Cruise Automation Chevy Bolt is using two Velodyne HDL-32E, each of which has a list price of $30,000. Price aside, Velodyne is really struggling to produce these units. Some recent HN comments have mentioned lead times of 40 weeks [1], though the upcoming Velodyne factory should alleviate these issues in 2018 [1].

We will need miracles in lidar technology for GM to hit their target of thousands of vehicles next year.

[1] https://news.ycombinator.com/item?id=13470401

Miracles in Lidar are coming. Quanergy, Infineon, Innoviz and Velodyne are all rushing to technologize solid state Lidar, which offers comparable performance specs to the spinning kfc bucket at quoted prices ranging from $250-$40. Imagine all the places we'll be using Lidar if it gets that cheap, the applications extend well beyond self driving cars.

From a capitalism point of view, the high price and long lead times imply that the market is signalling that it would like new entrants. Does Velodyne own patents that would keep out other players?

It will probably take longer than that. I've yet to see a non-tech demo fully self-driving car. If I'm being optimistic, I'd say 2020 at the very least for early models.

Isn't 2020 the definition of "a few years"?

I'm saying that GM will not have a fully self-driving car by 2018, so you wouldn't be able to take a nap in your car, even if you could afford it.

Reply on news.ycombinator.com