Learning Fast Adaptation with Meta Strategy Optimization

September 28, 2019 Β· Declared Dead Β· πŸ› IEEE Robotics and Automation Letters

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Wenhao Yu, Jie Tan, Yunfei Bai, Erwin Coumans, Sehoon Ha arXiv ID 1909.12995 Category cs.RO: Robotics Cross-listed cs.LG Citations 107 Venue IEEE Robotics and Automation Letters Last Checked 4 months ago
Abstract
The ability to walk in new scenarios is a key milestone on the path toward real-world applications of legged robots. In this work, we introduce Meta Strategy Optimization, a meta-learning algorithm for training policies with latent variable inputs that can quickly adapt to new scenarios with a handful of trials in the target environment. The key idea behind MSO is to expose the same adaptation process, Strategy Optimization (SO), to both the training and testing phases. This allows MSO to effectively learn locomotion skills as well as a latent space that is suitable for fast adaptation. We evaluate our method on a real quadruped robot and demonstrate successful adaptation in various scenarios, including sim-to-real transfer, walking with a weakened motor, or climbing up a slope. Furthermore, we quantitatively analyze the generalization capability of the trained policy in simulated environments. Both real and simulated experiments show that our method outperforms previous methods in adaptation to novel tasks.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Robotics

Died the same way β€” πŸ‘» Ghosted