@article{hu2022rac,
  author = {Edward S. Hu and Kun Huang and Oleh Rybkin and Dinesh Jayaraman},
  journal = {ICLR},
  title = {Know Thyself: Transferable Visuomotor Control Through Robot-Awareness},
  year = {2022},
  month = {Apr},
  pub_type = {conference},
  abstract = {Training visuomotor robot controllers from scratch on a new robot typically requires generating large amounts of robot-specific data. Could we leverage data previously collected on another robot to reduce or even completely remove this need for robot-specific data? We propose a "robot-aware" solution paradigm that exploits readily available robot "self-knowledge" such as proprioception, kinematics, and camera calibration to achieve this. First, we learn modular dynamics models that pair a transferable, robot-agnostic world dynamics module with a robot-specific, analytical robot dynamics module. Next, we set up visual planning costs that draw a distinction between the robot self and the world. Our experiments on tabletop manipulation tasks in simulation and on real robots demonstrate that these plug-in improvements dramatically boost the transferability of visuomotor controllers, even permitting zero-shot transfer onto new robots for the very first time.},
  url_pdf = {/publication/hu-2022-rac/hu-2022-rac.pdf},
  url_project = {https://hueds.github.io/rac/},
  url_code = {https://github.com/penn-pal-lab/robot_aware_control},
  url_openreview = {https://openreview.net/forum?id=o0ehFykKVtr},
  url = {https://arxiv.org/abs/2107.09047},
  url_arxiv = {https://arxiv.org/abs/2107.09047},
}