A Generalized Framework for Opening Doors and Drawers in Kitchen Environments

In this paper, we present a generalized framework for robustly operating previously unknown cabinets in kitchen environments. Our framework consists of the following four components: (1) a module for detecting both Lambertian and non-Lambertian (i.e. specular) handles, (2) a module for opening and closing novel cabinets using impedance control and for learning their kinematic models, (3) a module for storing and retrieving information about these objects in the map, and (4) a module for reliably operating cabinets of which the kinematic model is known. The presented work is the result of a collaboration of three PR2 beta sites. We rigorously evaluated our approach on 29 cabinets in five real kitchens located at our institutions. The paper is under review and available on demand. [ video ] [ code ]

Segmentation of Cluttered Scenes through Interactive Perception

This paper describes a novel object segmentation approach for autonomous service robots acting in human living environments. The proposed system allows a robot to effectively segment textured objects in cluttered scenes by leveraging its manipulation capabilities. In this interactive perception approach, 2D-features are tracked while the robot actively induces motions into a scene using its arm. Thereby, the robot autonomously infers appropriate arm movements which can effectively separate objects. The resulting tracked feature trajectories are assigned to their corresponding object by using a novel randomized clustering algorithm, which samples rigid motion hypotheses for the a priori unknown number of scene objects. We evaluated the approach on challenging scenes which included occluded and reflective objects, as well as objects of varying shapes and sizes. The paper is under review and available on demand. [ video ] [ code ]