Hacker News Reader

Comment by AlotOfReading | Hacker News Reader

As someone in the autonomous driving industry, maybe I can contribute a different perspective.

I totally agree that better designed cities would be superior. When I took the initiative to participate in my local government, I was screamed and cursed at for the most trivial improvements imaginable, like a shared bike lane on a few miles of a particularly dangerous street.

Multiply that by the opposition to actual improvements and advocacy in every single town, city, and state in America.

Fuck that. Self driving cars are way easier.

View on HN

Not totally on topic, but what's your take on being able to roll out autonomous vehicles in places with more "fluid" driving habits? I've driven in some crazy environments in the past, but I just came back from a very large country (let's call it an eighth of the world's population) where I can't imagine renting a car and driving there because of the near complete disregard they have for official traffic rules. Obviously they're just following a different set of rules, because people don't just crash into each other, but in that situation do you program the car to follow government rules or local custom?

Depends on what you mean and what you think driver behavior will be like. "Able to avoid hitting other things in a dense environment, but may not make forward progress" is fairly straightforward from the current state-of-the-art (though not necessarily easy!).

There's a reasonable argument to be made that making forward progress while remaining safe when intelligent agents are willing to play chicken requires a good theory of mind, but I can't formalize it. It's "just a hunch", but that seems like a Hard Problem to me and not necessarily something we'll completely solve soon. There's been work along these lines with statistical/ML methods, but my personal suspicion is that existing work would quickly fall apart in a truly hostile environment.

There's also the discussion around navigating non-roads, potholes, and other infrastructural issues that the industry hasn't solved yet. I don't think these are insurmountable, just difficult engineering issues.

The problem is much more widespread than just self driving cars. The US follows the model of making driving as predictable and event free as possible. Which results in people driving as fast as possible while paying the least attention. When someone does something unpredictable in the usually predicable environment, it results in death.

While its quite proven that slow, chaotic, and unpredictable environments are so much safer. Notice how walking through an all directions intersection crossing is incredibly chaotic, and yet no one collides with anyone else, or even when they do, it results in basically no harm.

> While its quite proven that slow, chaotic, and unpredictable environments are so much safer.

The "slow" part, for sure. Harm in a collision is largely a matter of kinetic energy.

The "chaotic and unpredictable" part, maybe not. From the statistics I've seen over the years, traffic-related death rates seem to be much higher in places with highly chaotic driving environments. For example, 140 annual traffic deaths per 100,000 vehicles in India vs. 14 in the USA.

Both India and China do have high rates of both traffic and pedestrian deaths compared to European countries.

I think the answer is going to be government rules because it's important that self-driving cars are predictable when there's no human driver you can have eye contact with in complex situations.

Reply on news.ycombinator.com