Autonomous door opening and plugging in with a personal robot

Abstract

Abstract-We describe an autonomous robotic system capable of navigating through an office environment, opening doors along the way, and plugging itself into electrical outlets to recharge as needed. We demonstrate through extensive experimentation that our robot executes these tasks reliably, without requiring any modification to the environment. We present robust detection algorithms for doors, door handles, and electrical plugs and sockets, combining vision and laser sensors. We show how to overcome the unavoidable shortcoming of perception by integrating compliant control into manipulation motions. We present a visual-differencing approach to highprecision plug-insertion that avoids the need for high-precision hand-eye calibration.

Door and cabinet recognition using Convolutional Neural Nets and real-time method fusion for handle detection and grasping

“…Kim et al [1] only detected 50% of the tested handles; Meeussen et af. [8] obtained around a 60% success rate in handle detection; Ruehr et af. [6] reached a stunning 93.3% of true positives using only depth data but had prior knowledge of the door and cabinet locations; Klingbeil et al [7] also reached 93.2% using only visual features.…”

“…These are not very relevant because the fast update of data ensures results are obtained almost every time step. The average processing time of the end-to-end system (except the grasping) is of l.8Hz; which is much faster than [5], since that approach does not detect handles, and [8], where just the data collection for one trial works at 1Hz.…”

“…To apply the following planar extraction, the full environments point cloud could be used, as done in [8], but that would use more information than necessary and defmitely slower the processing time.…”

“…[8] can easily be considered state-of-the-art due to its methods of detecting doors and handles and manipulating them; however the generation of the necessary data takes 10 seconds plus some extra computation time to finally be able to open a 978-1-5090-6088-7117/$31.00 ©20 17 IEEE 144…”

Improving robot manipulation with data-driven object-centric models of everyday forces

“…Recently, researchers have developed a number of robotic systems to operate doors between rooms (e.g., Klingbeil et al 2008;Jain and Kemp 2009a;Meeussen et al 2010;Chitta et al 2010;Kalakrishnan et al 2011;Kormushev et al 2011), and open cabinets, drawers, and appliances (e.g., Wieland et al 2009;Diankov et al 2008;Rühr et al 2012;Becker et al 2011). These efforts have focused on controllers and planners that enable a robot to open doors.…”