Learning Continuous Grasping Function With a Dexterous Hand From Human Demonstrations

We propose to learn generate grasping motion for manipulation with a dexterous hand using implicit functions. With continuous time inputs, the model can and smooth plan. name proposed Continuous Grasping Function (CGF). CGF is learned via generative modeling Conditional Variational Autoencoder 3D human demonstrations. will first convert large-scale human-object interaction trajectories robot demonstrations retargeting, then use these train CGF. During inference, we perform sampling different plans in simulator select successful ones transfer real robot. By training on diverse data, our allows generalization manipulate multiple objects. Compared previous planning algorithms, more efficient achieves significant improvement success rate when transferred Allegro Hand.