How to mitigate the challenges in the sim-to-real RL for a toy-sized humanoid robot?
To this end, we have proposed a RL framework leveraging observations from IMUs mounted on the feet.
Project Page: https://t.co/HGPR2OvD6D
arXiv: https://t.co/s3nmo2d7s7
Check out our #ICRA2025 paper where we train a robotic puppy to dance expressively!
Our method, Deep Fourier Mimic, a generalized version of DeepMimic, enables automatic parameterization of reference data.
Project site: https://t.co/HSBbQt7aZK
Video: https://t.co/yy3MTXOyIz
🧵