FRAM: Trajectory-Guided Visual Feature Selection for Compact Language-Conditioned Robot Manipulation
This work proposes the Future Representation Action Model (FRAM), a small policy that explicitly links the future end-effector trajectory to the current visual input and selects visual information based on future motion to obtain both high performance and robustness in a small robot policy.
Hiroshi Ito, Hyogo Hiruma, Yoshiki Kanai et al.
· 0 citations