Here is an approach that enables robots to perform unseen highly dynamic motions. This paper explores the idea of using a variational encoder to extract latent space encoding and train a conditional policy using the latent code. The approach allows a bipedal robot to execute highly dynamic moves. The below video from DisneyResearchHub shows how it works:
More like this ➡️ here
VMP: Versatile Motion Priors for Robustly Tracking Motion on Physical Characters
The robot can regain its balance when pushed.
[HT]
*Our articles may contain aff links. As an Amazon Associate we earn from qualifying purchases. Please read our disclaimer on how we fund this site.
