正在加载视频...

视频加载失败

Whatever happened to training physics based character controllers using RL? This was all the rage a few years ago. Every clip here looks cool af.

29,075 次观看 • 3 个月前 •via X (Twitter)

11 条评论

Chen Tessler @ CoRL 的头像
Chen Tessler @ CoRL3 个月前

still going strong. Check out our latest work GPC:

AVB 的头像
AVB3 个月前

This is really cool 😍

AVB 的头像
AVB3 个月前

Papers in this video:

Carlos DP ✈️ CoRL 2026 的头像
Carlos DP ✈️ CoRL 20263 个月前

it's still all the rage this year! Check out GPC, SMP, HIL, etc. These are experiments with GPC:

Chandra Shekhar 📟 的头像
Chandra Shekhar 📟3 个月前

peopel who were doing this are now busy minting synthetic data en masse

CIX 🦾 的头像
CIX 🦾3 个月前

We do this with robots with the @rek sim

pgBouncer 的头像
pgBouncer3 个月前

If a virtual object knows how to behave in accurately modeled virtual space, it should know how to in reality Then apply a multiverse simulation to train it on all possibly outcomes Insane potential. Something im very interested in

DK 的头像
DK3 个月前

RL slow, RL expensive, RL fails to generalize to edge cases etc

Fajar M Reza 的头像
Fajar M Reza3 个月前

Character-control progress may depend more on real-world physics than prettier demonstrations.

İsmail Keskin 的头像
İsmail Keskin3 个月前

Probably still eating too much resources. The people doing this research were mostly from AAA studios where resources are already tight. Also unlike traditional animation when this fails it fails in miserable ways. Indies could use this though.

Alec 的头像
Alec3 个月前

Been a bit busy, but wanted to jump back here and mention I'm working on exactly this. There's no reason this shouldn't be the future of characters in games.

相关视频