Show HN: Self-Parking Car Evolution

#1

Post by oleksiitwork » Tue, Sep 28, 2021, 3:23 PM UTC

Show HN: Self-Parking Car Evolution
trekhleb.dev

Re: Show HN: Self-Parking Car Evolution

#2

Post by drcongo » Tue, Sep 28, 2021, 4:08 PM UTC

Well it's outperforming Tesla FSD in generation 1.

Re: Show HN: Self-Parking Car Evolution

#3

Post by horsellama » Tue, Sep 28, 2021, 4:10 PM UTC

very cool, thanks for sharing!

all the steps are well described and the code can be used as a template for other examples

Re: Show HN: Self-Parking Car Evolution

#4

Post by SamBam » Tue, Sep 28, 2021, 4:14 PM UTC

This is cute, but am I missing some way to make it run faster? It seems to only happen in real time.

Group sizes of any more than about 10 make my machine slow to a crawl. Using the suggested population size of 500, then means it takes running 50 groups to pass 1 generation. With the suggested time of 17s, that's about 15 minutes per generation. The recommended 50 generations would therefore take over 12 hours.

And that's at the low-end of their recommendations. Population size of 1000 and 100 generations would take 50 hours.

I feel like usually with these things you want to run most of the generations headless, as fast as possible, and only show a few exemplars from each generation so the user can see how the evolution is progressing.

Re: Show HN: Self-Parking Car Evolution

#5

Post by Someone » Tue, Sep 28, 2021, 4:32 PM UTC

drcongo wrote: post #2

Well it's outperforming Tesla FSD in generation 1.

Having perfect sensors and only one task to do makes live a lot easier.

Re: Show HN: Self-Parking Car Evolution

#6

Post by Someone » Tue, Sep 28, 2021, 4:36 PM UTC

> Whenever the sensor doesn't see any obstacles it reports the value of 0. On the contrary, if the value of the sensor is small but not zero (i.e. 0.01m) it would mean that the obstacle is close.

I would guess that mapping ”nothing here” to a value that is higher than the others would give better results. The software wouldn’t have to learn that weird inversion where the safest value is very close to the least safe ones.

Re: Show HN: Self-Parking Car Evolution

#7

Post by trekhleb » Tue, Sep 28, 2021, 4:39 PM UTC

SamBam wrote: post #4

This is cute, but am I missing some way to make it run faster? It seems to only happen in real time. Group sizes of any more than about 10 make my machine slow to a crawl. Using the suggested population size of 500, then means it takes running 50 groups to pass 1 generation. With the suggested time of 17s, that's about 15 minutes per generation. The recommended 50 generations would therefore take over 12 hours. And t…

Yes, currently the simulation performance is one of the biggest issues (https://github.com/trekhleb/self-parking-car-evolution/issue...). You may try to check the “Performance boost” checkbox that simplifies the geometry. It should give you approx x1.5 performance increase. But even with x1.5 boost the performance is still an issue, yes

Re: Show HN: Self-Parking Car Evolution

#8

Post by riboflavin123 » Tue, Sep 28, 2021, 4:52 PM UTC

trekhleb wrote: post #7

SamBam wrote: post #4

This is cute, but am I missing some way to make it run faster? It seems to only happen in real time. Group sizes of any more than about 10 make my machine slow to a crawl. Using the suggested population size of 500, then means it takes running 50 groups to pass 1 generation. With the suggested time of 17s, that's about 15 minutes per generation. The recommended 50 generations would therefore take over 12 hours. And t…

Yes, currently the simulation performance is one of the biggest issues ( https://github.com/trekhleb/self-parking-car-evolution/issue... ). You may try to check the “Performance boost” checkbox that simplifies the geometry. It should give you approx x1.5 performance increase. But even with x1.5 boost the performance is still an issue, yes

The page says:

> For better results, increase the population size to 500-1000 and wait for 50-100 generations

Do you have example output from that? I'd expect my computer to take months of running 24/7 to get there.

Re: Show HN: Self-Parking Car Evolution

#9

Post by trekhleb » Tue, Sep 28, 2021, 4:58 PM UTC

riboflavin123 wrote: post #8

trekhleb wrote: post #7

Earlier quoted context omitted.

Yes, currently the simulation performance is one of the biggest issues ( https://github.com/trekhleb/self-parking-car-evolution/issue... ). You may try to check the “Performance boost” checkbox that simplifies the geometry. It should give you approx x1.5 performance increase. But even with x1.5 boost the performance is still an issue, yes

The page says: > For better results, increase the population size to 500-1000 and wait for 50-100 generations Do you have example output from that? I'd expect my computer to take months of running 24/7 to get there.

Yes, you may press the "Restore Evolution" button (at the bottom of the screen) and then upload the pre-saved training checkpoints from here https://github.com/trekhleb/self-parking-car-evolution/tree/...

You may also press the "Restore Evolution" button and then press "Use demo checkpoint" to use some pre-trained data.

Re: Show HN: Self-Parking Car Evolution

#10

Post by jetrink » Tue, Sep 28, 2021, 5:13 PM UTC

Someone wrote: post #6

> Whenever the sensor doesn't see any obstacles it reports the value of 0. On the contrary, if the value of the sensor is small but not zero (i.e. 0.01m) it would mean that the obstacle is close. I would guess that mapping ”nothing here” to a value that is higher than the others would give better results. The software wouldn’t have to learn that weird inversion where the safest value is very close to the least safe o…

Alternatively, it would be interesting to try having the sensors output 1/(distance + k), where k is maybe 0.1m. Then the output would naturally go to zero as things got further away.