Researchers are working towards a future where robots can assist people who need help getting dressed.

Data from the National Center for Health Statistics reveals that 92% of nursing facility residents and at-home care patients require assistance with dressing, a daily activity easily taken for granted by many.

“Remarkably, existing endeavors in robot-assisted dressing have primarily assumed dressing with a limited range of arm poses and with a single fixed garment, like a hospital gown,” says Yufei Wang, a Carnegie Mellon University’s Robotics Institute (RI) PhD student working on a robot-assisted dressing system.

“Developing a general system to address the diverse range of everyday clothing and varying motor function capabilities is our overarching objective. We also want to extend the system to individuals with different levels of constrained arm movement.”

The robot-assisted dressing system leverages the capabilities of artificial intelligence to accommodate various human body shapes, arm poses, and clothing selections. The team’s research used reinforcement learning—rewards for accomplishing certain tasks—to achieve their general dressing system. Specifically, the researchers gave the robot a positive reward each time it properly placed a garment further along a person’s arm. Through continued reinforcement, they increased the system’s learned-dressing strategy success rate.

The researchers used a simulation to teach the robot how to manipulate clothing and dress people. The team had to carefully deal with the properties of the clothing material when transferring the strategy learned in simulation to the real world.

“In the simulation phase, we employ deliberately randomized diverse clothing properties to guide the robot’s learned dressing strategy to encompass a broad spectrum of material attributes,” says Zhanyi Sun, a master’s student who also worked on the project. “We hope the randomly varied clothing properties in simulation encapsulate the garments’ property in the real world, so the dressing strategy learned in simulation environments can be seamlessly transferred to the real world.”

The researchers evaluated the robotic dressing system in a human study with 510 dressing trials across 17 participants with different body shapes, arm poses, and five garments. For most participants, the system was able to fully pull the sleeve of each garment onto their arm. When averaged over all test cases, the system dressed 86% of the length of the participants’ arms.

The researchers had to consider several challenges when designing their system. First, clothes are deformable in nature, making it difficult for the robot to perceive the full garment and predict where and how it will move.

“Clothes are different from rigid objects that enable state estimation, so we have to use a high-dimensional representation for deformable objects to allow the robot to perceive the current state of the clothes and how they interact with the human’s arm,” Wang says. “The representation we use is called a segmented point cloud. It represents the visible parts of the clothes as a set of points.”

Safe human-robot interaction was also crucial. It was important that the robot avoid both applying excessive force to the human arm and any other actions that could cause discomfort or compromise the individual’s safety. To mitigate these risks, the team rewarded the robot for gentle conduct.

Future research could head in several directions. For example, the team wants to expand the capabilities of the current system by enabling it to put a jacket on both of a person’s arms or pull a T-shirt over their head. Both tasks require more complex design and execution. The team also hopes to adapt to the human’s arm movements during the dressing process and to explore more advanced robot manipulation skills such as buttoning or zipping.

As the work progresses, the researchers intend to perform observational studies within nursing facilities to gain insight into the diverse needs of individuals and improvements that need to be made to their current assistive dressing system.

Wang and Sun recently presented their research at the Robotics: Science and Systems conference. The students are advised by Zackory Erickson, assistant professor in the RI and head of the Robotic Caregiving and Human Interaction (RCHI) Lab; and David Held, associate professor in the RI leading the Robots Perceiving And Doing (RPAD) research group.

Source: Carnegie Mellon University

Related

Robotic gripper is gentle enough to pick up a drop of water

Can touchy feely robots help you with laundry?

Robot ‘eyes’ aid people with profound motor impairments

An interface system with augmented reality technology could help people with profound motor impairments operate a humanoid robot to feed themselves and perform routine personal care tasks.

Those tasks might include feeding and performing routine personal care tasks such as scratching an itch and applying skin lotion.

The web-based interface displays a “robot’s eye view” of surroundings to help users interact with the world through the machine.

Described in PLOS ONE, the system could help make sophisticated robots more useful to people who don’t have experience operating complex robotic systems. Study participants interacted with the robot interface using standard assistive computer access technologies—such as eye trackers and head trackers—that they already used to control their personal computers.

The paper reports on two studies showing how such “robotic body surrogates”—which can perform tasks similar to those of humans—could improve the quality of life for users. The work could provide a foundation for developing faster and more capable assistive robots.

“Our results suggest that people with profound motor deficits can improve their quality of life using robotic body surrogates,” says first author Phillip Grice, a recent doctoral graduate of the Georgia Institute of Technology. “We have taken the first step toward making it possible for someone to purchase an appropriate type of robot, have it in their home, and derive real benefit from it.”

Water bottles and wash cloths

Grice and Charlie Kemp, professor in the biomedical engineering department at Georgia Tech and Emory University, used a PR2 mobile manipulator for the two studies. The wheeled robot has 20 degrees of freedom, with two arms and a “head,” giving it the ability to manipulate objects such as water bottles, washcloths, hairbrushes, and even an electric shaver.

“Our goal is to give people with limited use of their own bodies access to robotic bodies so they can interact with the world in new ways,” Kemp says.

In the first study, Grice and Kemp made the PR2 available across the internet to a group of 15 participants with severe motor impairments. The participants learned to control the robot remotely, using their own assistive equipment to operate a mouse cursor to perform a personal care task. Eighty percent of the participants could manipulate the robot to pick up a water bottle and bring it to the mouth of a mannequin.

“Compared to able-bodied persons, the capabilities of the robot are limited,” Grice says. “But the participants were able to perform tasks effectively and showed improvement on a clinical evaluation that measured their ability to manipulate objects compared to what they would have been able to do without the robot.”

User empowerment

In the second study, the researchers gave the PR2 and interface system to Henry Evans, a California man who has been helping Georgia Tech researchers study and improve assistive robotic systems since 2011.

“…he found new opportunities for using it that we had not anticipated.”

Evans, who has very limited control of his body, tested the robot in his home for seven days and not only completed tasks, but also devised novel uses combining the operation of both robot arms at the same time—using one arm to control a washcloth and the other to use a brush.

“The system was very liberating to me, in that it enabled me to independently manipulate my environment for the first time since my stroke,” Evans says. “With respect to other people, I was thrilled to see Phil get overwhelmingly positive results when he objectively tested the system with 15 other people.”

The way Evans developed new uses for the robot, combining motion of the two arms in ways they had not expected pleased the researchers, Grice says.

“When we gave Henry free access to the robot for a week, he found new opportunities for using it that we had not anticipated. This is important because a lot of the assistive technology available today is designed for very specific purposes.

“What Henry has shown is that this system is powerful in providing assistance and empowering users. The opportunities for this are potentially very broad.”

Universal design

The interface allowed Evans to care for himself in bed over an extended period of time. “The most helpful aspect of the interface system was that I could operate the robot completely independently, with only small head movements using an extremely intuitive graphical user interface,” he says.

The web-based interface shows users what the world looks like from cameras located in the robot’s head. Clickable controls overlaid on the view allow the users to move the robot around in a home or other environment and control the robot’s hands and arms.

When users move the robot’s head, for instance, the screen displays the mouse cursor as a pair of eyeballs to show where the robot will look when the user clicks. Clicking on a disc surrounding the robotic hands allows users to select a motion. While driving the robot around a room, lines following the cursor on the interface indicate the direction it will travel.

Building the interface around the actions of a simple single-button mouse allows people with a range of disabilities to use the interface without lengthy training sessions.

“Having an interface that individuals with a wide range of physical impairments can operate means we can provide access to a broad range of people, a form of universal design,” Grice notes.

“Because of its capability, this is a very complex system, so the challenge we had to overcome was to make it accessible to individuals who have very limited control of their own bodies.”

Robot surrogates

While the results of the study demonstrated what the researchers had set out to do, Kemp agrees they can still make improvements. The existing system is slow, and user mistakes can create significant setbacks. Still, he says, “People could use this technology today and really benefit from it.”

The developers will need to make significant reductions in cost and size to make the PR2 commercially viable, Evans says. The studies point the way to a new type of assistive technology, Kemp adds.

“It seems plausible to me based on this study that robotic body surrogates could provide significant benefits to users.”

The National Institute on Disability, Independent Living and Rehabilitation Research, the National Science Foundation, and the Residential Care Facilities for the Elderly of Fulton County funded the work. Willow Garage made the robot.

Kemp is a cofounder, a board member, an equity holder, and the CTO of Hello Robot Inc., which develops products related to this research. This research could affect his personal financial status. Georgia Tech has reviewed and approved the terms of this arrangement in accordance with its conflict of interest policies.

Source: Georgia Tech

Related

Star Wars-inspired robotic hand lets pianist play again

Watch: Robotics offer clues to fixing gait after stroke

Your future coworkers could be swarms of robots