Learning to jump high!
This video illustrates how a quadruped learns to jump as high as possible using Bayesian optimization with unknown constraints (BOC). Our results outperform manual tuning. The robot learns to (i) avoid failures and (ii) what is causing them. The former follows from BOC, while the latter is possible with a novel Gaussian process model for the constraint that learns the constraint threshold alongside with the constraint itself.
This is a complementary video to our RA-L/ICRA 2021 submission. Take a look to our preprint here https://arxiv.org/abs/2010.08669
Code: https://github.com/alonrot/classified_regression
First author's YouTube channel: https://www.youtube.com/channel/UC9YkR3t9NNUqsMs6ebaI46A
Видео Learning to jump high! канала Open Dynamic Robot Initiative
This is a complementary video to our RA-L/ICRA 2021 submission. Take a look to our preprint here https://arxiv.org/abs/2010.08669
Code: https://github.com/alonrot/classified_regression
First author's YouTube channel: https://www.youtube.com/channel/UC9YkR3t9NNUqsMs6ebaI46A
Видео Learning to jump high! канала Open Dynamic Robot Initiative
Показать
Комментарии отсутствуют
Информация о видео
23 октября 2020 г. 13:26:36
00:02:09
Другие видео канала
![Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motions](https://i.ytimg.com/vi/UmL4j0atcB8/default.jpg)
![Learning Agile Skills via Adversarial Imitation of Rough Partial Demonstrations](https://i.ytimg.com/vi/Pxq53vi7880/default.jpg)
![CASSI - Cooperative Adversarial Self-supervised Skill Imitation](https://i.ytimg.com/vi/3Q7gpUhnmUU/default.jpg)
![Controlling the Solo12 Quadruped Robot with Deep Reinforcement Learning](https://i.ytimg.com/vi/t-67qBxNyZI/default.jpg)
![SOLOBACKFLIP - Generating a full backflip](https://i.ytimg.com/vi/pp8UKMgnZ0A/default.jpg)
![SOLOLEAP - Moving forward with a jumping motion](https://i.ytimg.com/vi/YjK18ueJMhU/default.jpg)
![SOLOWAVE - Producing wave-like locomotion behavior](https://i.ytimg.com/vi/WzlOgAXqzVo/default.jpg)
![Improved Control Scheme for the Solo Quadruped and Experimental Comparison of MPCs](https://i.ytimg.com/vi/0MfbFre86nM/default.jpg)
![Real time contact planning and control of the Solo robot in 3d environments](https://i.ytimg.com/vi/gZprh_ofrfY/default.jpg)
![BiConMP: A Nonlinear Model Predictive Control Framework for Whole Body Motion Planning](https://i.ytimg.com/vi/dS_HUq1-pQY/default.jpg)
![Solo12 outdoor experiment at LAAS, Autonomy Update](https://i.ytimg.com/vi/kGpJpMjCQcA/default.jpg)
![First cordless test of Solo12, WiFi + Battery update](https://i.ytimg.com/vi/ZxmbFny51Pg/default.jpg)
![Solo8 v2 bounding based on Reinforcement Learning](https://i.ytimg.com/vi/4VvHGLNA3lk/default.jpg)
![Solo8 v2 jumping based on Reinforcement Learning](https://i.ytimg.com/vi/YJnbFJ-8Fkg/default.jpg)
![Solo12 Reactive Stepping in New York / NYU](https://i.ytimg.com/vi/1rJ3sFMBC1M/default.jpg)
![Hybrid Collision Model for Safety Collision Avoidance](https://i.ytimg.com/vi/fckpyq1DEqw/default.jpg)
![Learning a Centroidal Motion Planner for Legged Locomotion](https://i.ytimg.com/vi/vJ1VOUc8nH0/default.jpg)
![Reactive Walking Controller for Solo-12](https://i.ytimg.com/vi/VXCa4Ws_0Ik/default.jpg)
![A set of new tests on Bolt.](https://i.ytimg.com/vi/x2jYQdjT_es/default.jpg)
![First successful tests on our new biped robot "Bolt"](https://i.ytimg.com/vi/U8qYgGROyCM/default.jpg)