A new robotic system for the automated laundry

A new robotic system for the automated laundry

Credit: Bedada, Ahmadli and Palli.

Researchers at the University of Bologna and Electrolux recently developed a new robotic system that could assist humans in one of their most common daily chores, doing laundry. This system, introduced in an article published on SpringerLink’s Robotics friend of manhas been successfully trained in inserting garments and taking them out of the washing machine once the wash cycle has ended.

“Our recent project is the result of a collaboration between an industrial partner Electrolux and the University of Bologna,” Gianluca Palli, one of the researchers who conducted the study, told TechXplore. “There has been an interest from our industrial partner, Electrolux, to develop and implement robotic laundry handling in their home appliance test laboratory, to accelerate testing and evaluation of new products. This interest is motivated by the fact. that the implementation of robotic systems could potentially benefit by saving time and money by carrying out 24/7 testing and validation with little or no human intervention. “

The Automation and Robotics Laboratory (LAR) of the University of Bologna has conducted in-depth research focused on robotic manipulation. In recent years, researchers from this lab have introduced a number of promising systems, including robotic hand design and new methodologies for wiring switchgear and manipulating deformable objects.

“This correspondence interest between Electrolux and our lab led to a collaboration that was officially launched in 2017,” said Palli. “The main goal of this collaborative project, including our recent paper, was to develop and evaluate a robotic system (perception pipeline and control algorithms) to automate laundry operations.”

The goal of LAR and Electrolux’s recent research efforts is to eventually replace human operators in stores and laundry services with fully autonomous mobile robotic arms. This is far from an easy task, as clothes are highly deformable and robots typically struggle to efficiently handle deformable objects.

For a robotic system, doing laundry involves several steps. These steps include detecting which items need to be washed, picking them up, placing them in the washing machine, interpreting the information on the washing machine display, and adjusting the settings on the machine to perform the desired wash cycle.

A new robotic system for the automated laundry

Credit: Bedada, Ahmadli and Palli.

“Ensuring the correct insertion of the laundry and retrieval from the drum door region in the event that large items remain partially out of the washing machine is also critical to the success of the proposed system,” said Palli.

To improve their robotic system’s ability to grasp and manipulate clothing and other textiles, the researchers developed a point cloud-based perception algorithm. This algorithm can detect wrinkles on a fabric-based surface, calculate spline curves along the wrinkle-like structure, and estimate optimal grip frames.

“To improve the graspability of objects in the absence of wrinkles, we also evaluated a spot detection approach along with the ranking of the quality of the grip pose, which identifies the optimal poses,” said Palli. “We have also developed a washing machine user interface detection and interpretation algorithm based on deep learning, which allows us to fully automate the operation of the robotic laundry.”

After developing their robotic system and its various components, the researchers evaluated their effectiveness in a series of tests. In these tests, they used two robotic platforms they developed, namely the Tiago and Baxter robots, as well as an AEG washing machine.

“Our research shows that the problem of automated laundry operation has not been studied to its full extent in the literature,” Palli said. “This is the first time that the problem of fully robotic laundry cycle operation has been studied in all its aspects simultaneously, including gripping success, display recognition and robotic avoidance of collisions during interaction. , we were also able to develop and study a human-like tissue gripping strategy by targeting wrinkles and blemishes calculated from a live point cloud captured with the 3D Time of Flight (ToF) camera. “

In the future, the automated laundry system developed by this team of researchers could be implemented and tested in real contexts, to evaluate its performance in real contexts. While the system has shown promising results in initial tests, the researchers now intend to continue working on it to improve its generalizability and the ability to further manage textiles.

“There are two future research directions proposed,” Palli added. “The first line of extension is to evaluate deep neural networks for the task of detecting wrinkle identification. This will allow further generalization of the previously developed vision algorithms. The second area of ​​extension would be to fully implement the proposed system. in an industrial scenario and potentially quantify the economic and other benefits obtained from this robotic system. We know that our industrial partner will carry out further work in this direction. ”


Use the Rubik’s Cube to improve and evaluate robot manipulation


More information:
Wendwosen B. Bedada et al, Robotic laundry handling with interpretation of the appliance user interface, Robotics as a friend of man 2021 (2022). DOI: 10.1007 / 978-3-030-96359-0_7

© 2022 Science Network X

Quote: A new robotic automated laundry system (2022, May 10) recovered May 10, 2022 from https://techxplore.com/news/2022-05-robotic-automated-laundry.html

This document is subject to copyright. Outside of any commercial fairness for private study or research purposes, no part may be reproduced without written permission. The content is provided for informational purposes only.

Leave a Comment

Your email address will not be published.