HERB: a home exploring robotic butler

We describe the architecture, algorithms, and experiments with HERB, an autonomous mobile manipulator that performs useful manipulation tasks in the home. We present new algorithms for searching for objects, learning to navigate in cluttered dynamic indoor scenes, recognizing and registering objects accurately in high clutter using vision, manipulating doors and other constrained objects using caging grasps, grasp planning and execution in clutter, and manipulation on pose and torque constraint manifolds. We also present numerous severe real-world test results from S.S. Srinivasa ( ) · D. Ferguson · C.J. Helfrich Intel Research Pittsburgh, 4720 Forbes Avenue, Suite 410, Pittsburgh, PA 15213, USA e-mail: [email protected] C.J. Helfrich e-mail: [email protected] D. Berenson · A. Collet · R. Diankov · G. Gallagher · G. Hollinger · J. Kuffner · M.V. Weghe The Robotics Institute, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA D. Berenson e-mail: [email protected] A. Collet e-mail: [email protected] R. Diankov e-mail: [email protected] G. Gallagher e-mail: [email protected] G. Hollinger e-mail: [email protected] J. Kuffner e-mail: [email protected] M.V. Weghe e-mail: [email protected] the integration of these algorithms into a single mobile manipulator.