Poster
Fourier Transporter: Bi-Equivariant Robotic Manipulation in 3D
Haojie Huang · Owen Howell · Dian Wang · Xupeng Zhu · Robert Platt · Robin Walters
Halle B #34
Abstract:
Many complex robotic manipulation tasks can be decomposed as a sequence of pick and place actions. Training a robotic agent to learn this sequence over many different starting conditions typically requires many iterations or demonstrations, especially in 3D environments. In this work, we propose Fourier Transporter (), which leverages the two-fold symmetry in the pick-place problem to achieve much higher sample efficiency. is an open-loop behavior cloning method trained using expert demonstrations to predict pick-place actions on new configurations. is constrained by the symmetries of the pick and place actions independently. Our method utilizes a fiber space Fourier transformation that allows for memory-efficient computation. Tests on the RLbench benchmark achieve state-of-the-art results across various tasks.
Chat is not available.