Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning
Nvidia presented parts of this work at GTC 2022, revealing our humanoid-quadruped transformer!
Title: Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning
Authors: Eric Vollenweider, Marko Bjelonic, Victor Klemm, Nikita Rudin, Joonho Lee and Marco Hutter
Paper submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems in Kyoto.
Preprint:
Abstract: In recent years, reinforcement learning (RL) has shown outstanding performance for locomotion control of highly articulated robotic systems. Such approaches typically involve tedious reward function tuning to achieve the desired motion style. Imitation learning approaches such as adversarial motion priors aim to reduce this problem by encouraging a pre-defined motion style. In this work, we present an approach to augment the concept of adversarial motion prior-based RL to allow for multiple, discretely switchable styles. We show that multiple styles and skills can be learned simultaneously without notable performance differences, even in combination with motion data-free skills. Our approach is validated in several real-world experiments with a wheeled-legged quadruped robot showing skills learned from existing RL controllers and trajectory optimization, such as ducking and walking, and novel skills such as switching between a quadrupedal and humanoid configuration. For the latter skill, the robot is required to stand up, navigate on two wheels, and sit down. Instead of tuning the sit-down motion, we verify that a reverse playback of the stand-up movement helps the robot discover feasible sit-down behaviors and avoids tedious reward function tuning.
Note: The following parts of the video are sped up:
- Door opening and closing when the robot stands inside the elevator between 00:20 and 00:21 ( 200%)
- In-between the standing up and sitting down sequence between 00:49 and 01:10 ( 200% only the navigation on two legs)
- Reaction with the crowd between 1:41 and 1:50 ( 150%)
- Last drone footage after 2:01 ( 200%)
Learn more at
1 view
276
81
2 weeks ago 00:31:07 1
35 Advanced English Verb Rules
2 weeks ago 00:00:22 1
Do it! Other ways to say Daily English speaking practice English conversation
1 month ago 00:04:52 1
“Another Brick in the Wall Pt. 2“ (PINK FLOYD) - Solo Fingerstyle w/ GUITAR TABS
1 month ago 00:08:41 1
HOW TO DEEP THROAT | Get Her to Deep Throat You & Give You Epic Blow Jobs
1 month ago 00:15:16 1
Normal English VS Native English
1 month ago 00:40:42 1
How the Untalented Become Talented Through a Deal with the Devil Shaytan
1 month ago 00:01:48 1
Evertried - Release Date Trailer | PS4
1 month ago 00:00:43 1
Normal English vs Advanced English #english #shortfeed #viralshorts
1 month ago 00:01:27 3
glow dm skills android handcam
1 month ago 00:15:58 1
AMD Radeon RX 560 4GB: gameplay в 25 популярных играх при Full HD
1 month ago 00:01:00 1
MOTIVATIONAL : Best VLSI Training in INDIA | 100%Job Assistance | Job Oriented Advanced VLSI Courses
1 month ago 00:12:14 1
English Speaking Skills - How Would You Do Things - Advanced English Daily Conversation Practice
1 month ago 00:52:00 1
CBS’ Tracker: Justin Hartley in Conversation with Tommy DiDario
1 month ago 00:00:00 1
Portrait Drawing Tutorial
1 month ago 00:00:17 1
Epic Hot Three Section Staff Showdown! #kungfu #martialarts