Goal-conditioned dual-action deep imitation learning for banana peeling

The proposed goal-conditioned dual-action (GC-DA) deep imitation learning model was tested with the banana peeling task in the real dual-arm robot manipulator (UR5).

BibTex: