Terrain Traversal with Reinforcement Learning | Two Minute Papers #26
238Reinforcement learning is a technique that can learn how to play computer games, or any kind of activity that requires a sequence of actions. In this case, we would like a digital dog to run, and leap over and onto obstacles by choosing the optimal next action. It is quite difficult as there are a lot of body parts to control in harmony. And what is really amazing is that if it has learned everything properly, it will come up with exactly the same movements as we'd expect animals to do in real life! In this technique, dogs were used to demonstrate that reinforcement learning works well in this context, but it's worth noting that it also works with bipeds.
_____________________________
The paper "Dynamic Terrain Traversal Skills Using Reinforcement Learning " is available here:
http://www.cs.ubc.ca/~van/papers/2015-TOG-terrainRL/
Recommended for you:
Digital Creatures Learn To Walk - https://www.youtube.com/watch?v=kQ2bqz3HPJE
Subscribe if you would like to see more of these! - http://www.youtube.com/subscription_center?add_user=keeroyz
Thumbnail image by localpups (CC BY 2.0). It was slightly edited (flipped, color adjustments, content aware filling) - https://flic.kr/p/wXfFt1
Splash screen/thumbnail design: Felícia Fehér - http://felicia.hu
Károly Zsolnai-Fehér's links:
Patreon → https://www.patreon.com/TwoMinutePapers
Facebook → https://www.facebook.com/TwoMinutePapers/
Twitter → https://twitter.com/karoly_zsolnai
Web → https://cg.tuwien.ac.at/~zsolnai/