RVT-2: Learning Precise Manipulation from Few Demonstrations

In this work, we study how to build a robotic system that can solve multiple3D manipulation tasks given language instructions. To be useful in industrialand household domains, such a system should be capable of learning new taskswith few demonstrations and solving them precisely. Prior works, like PerActand RVT, have studied this problem, however, they often struggle with tasksrequiring high precision. We study how to make them more effective, precise,and fast. Using a combination of architectural and system-level improvements,we propose RVT-2, a multitask 3D manipulation model that is 6X faster intraining and 2X faster in inference than its predecessor RVT. RVT-2 achieves anew state-of-the-art on RLBench, improving the success rate from 65% to 82%.RVT-2 is also effective in the real world, where it can learn tasks requiringhigh precision, like picking up and inserting plugs, with just 10demonstrations. Visual results, code, and trained model are provided at:https://robotic-view-transformer-2.github.io/.