> To select which ones were best, each variation got assigned a ‘fitness’ score based on its walking speed, clearance and material use. I also rewarded extra points to variants that had leg tips which moved more horizontally and more smoothly during the lowest third of their cycle to make it favor solutions in which a group of three legs would work together to minimize bobbing and foot slip.
I am not sure what approach was used here, it sounds like RL or maybe just simulations. I am not sure I’d say this is AI.
There’s no clear definition currently, but based on my work in IEEE standards and my own research, I always define AI as an autonomous system which determines paths to its desired outcomes using multimodal inputs.
While I think the computation system described in this post is very cool and sophisticated, I liken it more to a computation engine. It seems to me that computation engines follow an algorithm to determine optimal paths to a previously determined and fixed desired outcome, but they don’t update the outcome goal based on new information.
If the computation engine described in this post could use existing information to form its own goal or idea of a desired outcome, then I’d say it’s AI.
I’d love to do a blog post on this idea because I am sure many might have good reasons to disagree with it. It would be cool if someone else does it, I’d love to provide feedback as I don’t have the time for longform writing.
> To select which ones were best, each variation got assigned a ‘fitness’ score based on its walking speed, clearance and material use. I also rewarded extra points to variants that had leg tips which moved more horizontally and more smoothly during the lowest third of their cycle to make it favor solutions in which a group of three legs would work together to minimize bobbing and foot slip.
I am not sure what approach was used here, it sounds like RL or maybe just simulations. I am not sure I’d say this is AI.