0
Select Articles

Robots for Infants PUBLIC ACCESS

Some Special-Needs Babies Lead Lives with Limited Mobility. Engineers have Begun to Build Devices that Allow these Children to Move About Independently.

[+] Author Notes

Sunil K. Agrawal is a professor of mechanical engineering at the University of Delaware in Newark. He is an ASME Fellow and in 2005 served as a conference chair for the ASME Mechanisms and Robotics Conference.

Mechanical Engineering 133(03), 50-51 (Mar 01, 2011) (2 pages) doi:10.1115/1.2011-MAR-7

Abstract

This article reviews the role of robots in special-needs infants. Motorized wheelchairs can help provide some measure of mobility to these special-needs infants. The pediatric crawlers support the weight of the infant’s torso to make it easier for the arms and legs to provide propulsion. The use of kick- and joystick-controlled robotic devices provides an opportunity for special-needs infants to make perceptual, cognitive and social development gains at what would be considered the appropriate age. The Mechanical Systems Laboratory at the University of Delaware has been exploring means to provide real mobility to infants and toddlers with special needs. The results have been promising: the children have intuitively figured out how to use their legs to make the robot move and how to turn the robot via the joystick.

Beginning at about six months of age, most infants begin to explore the world independent of others. They do this mostly by crawling. In the process of independent locomotion, infants undergo major changes in perception, cognition, and social and emotional development. Studies have shown that the rhythmic pattern involved in moving arms and legs in a coordinated manner impacts the way the developing brain grows and the ease with which it learns. In many ways, crawling is what helps turn babies into children.

Some infants, however, are unable to crawl. Infants with special conditions such cerebral palsy can have muscles so weak or coordination so poor that the onset of crawling is delayed by many months or even longer. Because of this, many important brain functions are slow to develop in such infants.

Motorized wheelchairs can help provide some measure of mobility to these special-needs infants. But currently, they are almost never used by infants less than 20 months old, because of safety concerns, and doctors tend to discourage motorized chairs for children under five. Children ought to be mobile well before then.

A less daunting piece of equipment is the pediatric crawler. This is a platform or suspended support sling mounted on casters. The crawlers support the weight of the infant's torso to make it easier for the arms and legs to provide propulsion. To be sure, such crawlers help in self-propulsion, but they are not suitable for all infants with mobility problems. The pediatric crawler enables the infant to move as far and as fast as his muscles and coordination will allow, and for many special needs babies, that isn’t much.

ROBO-CRAWLER: A camera tracks the feet and as they kick, it goes.

Grahic Jump LocationROBO-CRAWLER: A camera tracks the feet and as they kick, it goes.

The Mechanical Systems Laboratory at the University of Delaware has been exploring means to provide real mobility to infants and toddlers with special needs. Cole Galloway, a professor with expertise in infant motor learning, and I have studied how the use of robotics and sensors can create safe, yet powerful medical devices that can enable physically disabled children to explore the world around them and, in doing so, develop important social and cognitive capacities.

One of the experimental devices we developed recently is intended to translate relatively weak and uncoordinated arm and leg motions into strong, directed motion. Our preliminary results suggest to us that this experimental crawler may one day lead to a device that will enable certain special-needs children to receive many of the benefits from crawling that other children get.

It may not seem obvious that improving mobility will result in cognitive growth in infants, but we have seen it in our lab. In 2009, we reported on work with a seven-month-old child, Andrew, who was born with spina bifida. Spina bifida is a neural tube defect involving an incomplete closure of the spine. In some cases, like Andrew's, the spinal cord actually protrudes through the opening in the spine. For Andrew, the likelihood was that he would not walk before the age of two, and then only with difficulty, and there was a greater than 25 percent chance that he would use a wheelchair his entire life.

In the lab, Andrew was trained to operate a joystick-driven robot—the iRobot MagellanPro—that had been fitted with a commercial infant booster seat. During three or four sessions a week from seven months to 12 months of age, Andrew learned how to operate the joystick and to turn the movements of the joystick into locomotion in the robot. During some of the training, he would drive the robot six feet or so to retrieve a toy from his mother or a researcher. Other times, he was allowed to explore the experimental area, which was part of a gymnasium on the university campus.

After those five months of work, Andrew's success at driving the robot showed considerable improvement. More interesting was the change in scores on tests measuring cognitive and verbal skills. At the beginning of the training, Andrew scored one to two months behind the average for a seven-month-old infant. When he was given the same tests at the end of the experiment, Andrew's scores had caught up with his chronological age (12 months at that point) and had actually exceeded it in some measures. The newfound ability to explore the world with some degree of independence seemed to a provide real cognitive boost.

Of course, a purely joystick-driven robot develops only one set of motor skills. To better capture the coordination needed for crawling, it would be necessary to have a device that places the infant in a more-or-less horizontal position and uses the movements of both the arms and the legs to control the robotic motion.

Once one makes that design decision, other requirements manifest themselves. First and foremost is safety and child-friendliness: infants using the experimental device must be both secure and comfortable. The device must also have an intuitive interface connecting the motion of the arms and legs to the motion of the robot, and provide unrestricted movement of the limbs to enable the opportunity for exercise.

With those constraints in mind, we considered several interface combinations. One potential method would strap accelerometers to the infant's arms and legs; as the child moved his limbs, the signals from the sensors would direct the robotic platform. Unfortunately, it was difficult to accurately translate those signals into directional information.

Another potential solution would involve placing strain gauges under the board that supports the infant's belly and correlate the torso motions to crawling. Such a solution would be relatively inexpensive and would not require fine motor skills on the part of the child, but the system to acquire the data and convert it into computer code is not readily available.

BOOSTER SEAT DRIVER: Operating a robot like this can provide a measureable cognitive gain.

Grahic Jump LocationBOOSTER SEAT DRIVER: Operating a robot like this can provide a measureable cognitive gain.

Instead, we developed a hybrid control system. Locomotion is controlled by the legs through use of positional markers attached to the infant's ankles. As the child kicks, the motion of the markers is picked up by a Web camera mounted above the legs; software converts velocity of the markers to a velocity for the robotic platform. This creates an intuitive control, since the faster the child kicks, the faster the robot goes.

Direction is determined through the activation of a commercial joystick. As the infant pushes the joystick one way or another, the robot turns.

The subject infant rested on a platform and was secured in place by a seatbelt and a foam wedge. The platform sat atop a three-wheeled robot from MobileRobots of Amherst, N.H. Because the robot was relatively squat, the infant was never more than a foot off the floor during the experiments.

Our preliminary tests centered on the feasibility of the device as a means of transportation, and as such, we used children who were developing typically. Thus far, the results have been promising: the children have intuitively figured out how to use their legs to make the robot move and how to turn the robot via the joystick.

The children who have used this system so far already knew how to crawl, so they were familiar with the idea of moving their legs to go places. The real test is to begin studies using special-needs infants who have never crawled. How readily will they come to understand that swinging their legs back and forth translates into purposeful movement? And will learning to use a device such as this one provide the cognitive boost that infants gain from crawling?

To be clear, this sort of motorized mobility for infants is not intended to replace efforts to teach special-needs infants how to walk upright with the assistance of equipment. That work, if successful, obviously provides the best future for such children. But it can take years for children to achieve that goal, and in the meantime they are missing out on important brain development opportunities that are gained through self-guided exploration.

The use of kick- and joystick-controlled robotic devices provides an opportunity for special needs infants to make perceptual, cognitive and social development gains at what would be considered the appropriate age.

References

Chen, Liang, Dolph, Ragonesi, Galloway, and Agrawal. (2010) “Design of a Novel Mobility Interface for Infants on a Mobile Robot by Kicking.” Journal of Medical Devices 4 (3): 031006.1-031006.5
Lynch, Rhu, Agrawal, and Galloway. (2009) “Power Mobility Training for a 7-month-old Infant with Spina Bifida.” Pediatric Physical Therapy 21 (4): 362– 368 [CrossRef] [PubMed]

Cite this article as: Sunil K. Agrawal, Robots for Infants: Some special-needs babies lead lives with limited mobility. Engineers have begun to build devices that allow these children to move about independently., Renew. Energy Environ. Sustain. 00, 00 (0000)

Copyright © 2011 by ASME
View article in PDF format.

References

Chen, Liang, Dolph, Ragonesi, Galloway, and Agrawal. (2010) “Design of a Novel Mobility Interface for Infants on a Mobile Robot by Kicking.” Journal of Medical Devices 4 (3): 031006.1-031006.5
Lynch, Rhu, Agrawal, and Galloway. (2009) “Power Mobility Training for a 7-month-old Infant with Spina Bifida.” Pediatric Physical Therapy 21 (4): 362– 368 [CrossRef] [PubMed]

Figures

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In