Estimator-Coupled Reinforcement Learning for Robust Purely Tactile In-Hand Manipulation

no code implementations • 7 Nov 2023 • Lennart Röstel, Johannes Pitz, Leon Sievers, Berthold Bäuml

Hence, separately training the controller and the estimator and combining the two at test time leads to poor performance.

reinforcement-learning

Combining Shape Completion and Grasp Prediction for Fast and Versatile Grasping with a Multi-Fingered Hand

no code implementations • 31 Oct 2023 • Matthias Humt, Dominik Winkelbauer, Ulrich Hillenbrand, Berthold Bäuml

We present a novel, fast, and high fidelity deep learning pipeline consisting of a shape completion module that is based on a single depth image, and followed by a grasp predictor that is based on the predicted object shape.