Cited on HN 6 times, 2014–2021, never outright recommendedscore 0
Years with an HN recommendation, 2007–2026
HN commenters cite it as an example of a hard, chaotic control problem that fast feedback and good control design can solve.
A TU Wien control-systems demo in which a triple pendulum on a cart is swung up and stabilized using constrained feedforward plus optimal feedback control. It accompanies a 2013 Automatica paper.
“Triple!!! inverted pendulum balancing with a physical cart (2011), a much much much harder task” (HN, 2016)
Cool. Feature request: mount the pivot on a programmable cart. It'd be fun to write software to balance the double pendulum. e.g. [link] and, balancing a triple pendulum: [this video]
Two-pole balancing (2009): [link] Double inverted pendulum balancing (2015), a much harder task: [link] Double inverted pendulum balancing with a physical cart (2011), a much much harder task: [link] Triple!!! inverted pendulum balancing with a physical cart (2011), a much much much harder task: [this video]
Demo of balancing a triple inverted pendulum on a physical cart.
The single inverted pendulum balancing problem has been solved using neural nets, fuzzy logic, and nonlinear control theory. It's a standard problem in controls classes. Here's a system learning how to do this.[1] Takes about 200 trials. Here's the triple inverted pendulum balancing problem, solved using feedforward control.[2] [1] [link] [2] [this video]
Triple inverted pendulum balanced using feedforward control.
When you have mutliple vehicles, each potentially with their own variant of control software, then each could be internally deterministic but "foreign" to the other vehicle. Kind of like a double pendulum. Edit: But also like that cool double/triple pendulum thing ( [this video] ) you don't necessarily have to be able to predict the future if you are simply monitoring the present state at a high enough sampling rate. Note that this thing is sampling at 1ms. Not sure what kind of sampling would be necessary for cars, certainly a function of velocity and braking power.
Fast-sampling feedback control stabilizes a chaotic triple pendulum.
Double and triple pendulums have been solved: [this video] I have little doubt cancer will be someday as well. It might not -- probably won't -- happen in my lifetime. But if humanity hasn't collapsed, I find it very unlikely that we're still dealing with cancer in a thousand years.
Shows a triple pendulum stabilized, cited as a solved hard problem.
> We don't know how to make AI that will learn to play even simple game like montezuma's revenge in some reasonable play time like 30 min. Explain to me why a Tool-assisted Speedrun doesn't count as "AI". Its certainly not a human who is playing that ( [link] ). > Something that could in one package: I'm not sure if that sort of thing is worth the investment. But if we have "Fly the F35 airframe" (which is incredibly unstable), then we suddenly reach "only computers can fly this thing" levels. Just the assumption of super-human control systems is baked into modern products. [this video] Our cell phones are largely manufactured by machines these days. A few tidbits here and there remain in human hands for cheapness... but a lot of today's stuff is basically an AI. [link]
Example of superhuman machine control of an unstable system.
Video list and Hacker News comment labels from HackerTube (prompted by @karpathy), released under ODC-By 1.0, built on the open-index/hacker-news archive. Hacker News data is provided by Y Combinator; comments remain the work of their authors. Video titles, channel names and thumbnails belong to their creators and are shown via YouTube. HackerTube Watch is an unofficial edition. Data export of 2026-10-31.
Plays via YouTube's privacy-enhanced embed
6 Hacker News comments
From the HN archive; up to 20 kept per video, favoring different commenters. Recommendations first.
Read more