Robotic cleaning, Maryland Robotics Center, University of Maryland, College Park, USA

[Airicist]

Developer - Maryland Robotics Center

[Airicist]

Bimanual robotic cleaning of compliant parts

Published on Dec 23, 2015

This video demonstrates an approach to automatically learn a bimanual robotic cleaning task on compliant objects. One robot grasps the object, while the other robot cleans it. Given a part with unknown deformation characteristics, the system visually detects the regions to be cleaned, and generates plans for both the graspingand cleaning arms. As the system performs cleaning attempts and gains experience with multiple new parts, it learns models of the part deformation depending on the cleaning force and grasping parameters. A planner iteratively generates tool paths for both robots using the available knowledge to optimize the cleaning time, including (1) delays from regrasping a part to minimize deflection and (2) time taken for repeated cleaning attempts over regions that remained dirty. A nonparametric deflection model is learned separately for each part, with minimal assumptions of the material behavior. We demonstrate the approach on a system of two KUKA LWR iiwa robots and a set of thin planar parts. Results indicate that the system is effective at rapidly learning part deformation models to enable effective iterative cleaning performance.

[Airicist]

Robotic cleaning by scrubbing

Published on Dec 24, 2015

This video demonstrates an approach for robotic cleaning by scrubbing. Planning algorithms are developed for cleaning stains on a curved object. Removing the stain may require multiple reorientations of the part and some portions of the stain may require multiple cleaning passes. The experimental setup involves two robot arms. The first arm immobilizes the object. The second arm moves the cleaning tool. The algorithm analyzes the stain and determines the sequence of orientations needed to clean the part based on the kinematic constraints of the robot arm. Each orientation is called a cleaning setup. Our algorithm uses a depth-first branch-andbound search to generate setup plan solutions. We also compute the cleaning trajectories and select the cleaning parameters to maximize the cleaning performance. The algorithm generates multi-pass trajectories by replanning based on the observed cleaning performance. Numerical simulations and cleaning experiments with two Kuka robots are used to validate our approach.

[Airicist]

Compliant object cleaning

Published on Feb 2, 2016

This video demonstrates a robotic system for cleaning stains on flexible and fragile objects like a plastic manikin. The perception and planning modules are described. The ability of the system to replan and adapt to unexpected motions of the object is demonstrated.