Learning Multi-step Robotic Manipulation Policies from Visual Observation of Scene and Q-value Predictions of Previous Action.

Sulabh Kumra, Shirin Joshi, Ferat Sahin

Browse the full ICRA paper archive.