NSF Convergence Accelerator Track M: Soft Growing Robots for Mobility Support

In the post-pandemic era, the US faces an eldercare crisis. A severe workforce shortage, combined with the high costs associated with eldercare, is forcing nursing homes to close or provide substandard care. Furthermore, over 700,000 older adults are on waiting lists for Medicaid home and community-based services provided by individual states. Outside of professional caregiving, care responsibilities often fall to family members, relatives, and friends. Over 53 million private caregivers in the US are experiencing enormous physical, mental, and financial stress caring for older adults; many reduce their working hours and even give up their full-time jobs. This project aims to assist both older adults and caregivers in laborious and even dangerous mobility support. Transferring an older adult from a bed to a wheelchair, for example, is a fatiguing, strenuous task, causing numerous caregivers back injuries. An easy-to-use system of gentle yet strong robotic belts that go beneath the back of older adults and lift their body could safely transfer them to a wheelchair. No need to call a caregiver. Older adults can get out of bed on their own and live more independently. The special belt design was inspired from biological species with soft growing appendages such as marine worms and climbing vanes. Extending a tip underneath a heavy human body is similar to burrowing into soil. Some ringed worms evert their internal organs to act like a wedge, traveling through sediment by crack propagation. This marine worm’s eversion mechanism has been realized with a soft robotic device consisting of a double-layered tubular sheet. As pressurized air is supplied to the tubular sheet, the tip extends and goes underneath a human body. The tubular sheet can have a high tensile strength to lift the body. This project will establish the soft extendable belt technology for holding and lifting a human and develop a functional prototype for safe, gentle, and autonomous transfer of older adults. Successful deployment of support equipment cannot be achieved by advanced design alone. Product development methodologies will be applied to obtain feedback from end users throughout the project, and strategies will be established for effectively deploying the technology and removing barriers for adoption. Particularly important is engagement and guidance of the users, so that they can use the equipment easily and safely. This project will apply a Large Language Model and human-robot interaction methods to enhance user engagement and guidance. This project will work closely with care professionals at nursing facilities. Data of conversations and physical interactions between older adults and caregivers, as well as with the care robot, will be acquired and used for generating an AI-based user guidance and interaction system. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.