Skip to content

Commit

Permalink
UPdate
Browse files Browse the repository at this point in the history
  • Loading branch information
Skylark0924 committed Dec 25, 2024
1 parent a711720 commit fdd7365
Showing 1 changed file with 2 additions and 27 deletions.
29 changes: 2 additions & 27 deletions doc/source/lfd/RofuncRL/HOTU.md
Original file line number Diff line number Diff line change
@@ -1,28 +1,3 @@
# Human-Humanoid Robots Cross-Embodiment Behavior-Skill Transfer Using Decomposed Adversarial Learning from Demonstration

Due to the large number of GIFs on this page (though I've tried my best to compress and speed up these GIFs), loading may take a bit longer—**thank you for your patience**.

You can either view the [GitHub page](https://github.com/Skylark0924/Rofunc/blob/main/doc/source/lfd/RofuncRL/HOTU.md) or the precompiled [document](https://rofunc.readthedocs.io/en/latest/lfd/RofuncRL/HOTU.html). Please feel free to decide based on the loading speed.

## Kinematic Motion Retargeting

### Speed: 4x

<img src="../../../img/task_gif/UDH_Random_Motion.gif" width=25% /><img src="../../../img/task_gif/H1_Random_Motion.gif" width=25% /><img src="../../../img/task_gif/Walker_Random_Motion.gif" width=25% /><img src="../../../img/task_gif/Bruce_Random_Motion.gif" width=25% />

<img src="../../../img/task_gif/UDH_wave.gif" width=25% /><img src="../../../img/task_gif/H1_wave.gif" width=25% /><img src="../../../img/task_gif/Walker_wave.gif" width=25% /><img src="../../../img/task_gif/Bruce_wave.gif" width=25% />

## Perturb after Behavior Primitive Pre-training

### Speed: 1x

<img src="../../../img/task_gif/UDH_perturb.gif" width=50% /><img src="../../../img/task_gif/H1_perturb.gif" width=50% /><img src="../../../img/task_gif/NAVIAI_perturb.gif" width=50% /><img src="../../../img/task_gif/Bruce_perturb.gif" width=50% />

<img src="../../../img/task_gif/UDH_perturb2.gif" width=100% /><img src="../../../img/task_gif/H1_perturb2.gif" width=100% /><img src="../../../img/task_gif/NAVIAI_perturb2.gif" width=100% /><img src="../../../img/task_gif/Bruce_perturb2.gif" width=100% />

## Box Carrying Task (from random starting position to random target position with different box sizes)

### Speed: 4x

<img src="../../../img/task_gif/UDH_multi_motion.gif" width=100% /><img src="../../../img/task_gif/H1_multi_motion.gif" width=100% /><img src="../../../img/task_gif/NAVIAI_multi_motion.gif" width=100% />
# Human-Humanoid Robots Cross-Embodiment Behavior-Skill Transfer Using Decomposed Adversarial Learning from Demonstration

This page is reconstructing (due to the video format), will be back soon after Christmas.

0 comments on commit fdd7365

Please sign in to comment.