Hello Robot Receives Nearly $3M NIH Grant for Assistive Robotics Research

Hello Robot has received nearly $3 million in federal funding from the National Institutes of Health to develop and test its Stretch mobile manipulator as an assistive robot for older adults with mild cognitive impairment and early Alzheimer's disease. This money is not going toward a general-purpose robot in a laboratory, but is explicitly aimed at a specific population and a specific setting: older adults whose cognition is declining, and the homes and care facilities where they live.

The project runs for three years, is a collaboration between Hello Robot and the University of Illinois Urbana-Champaign, and is conducted through the National Institute on Aging under the NIH. The relationships among funder, executor, and academic partner are clear: federal money flows through the aging-research channel to a joint project between a robotics company and a university.

According to the company, the project began on September 22 and is expected to continue until May 2029. These start and end dates come from the company's account, and the report provides no independent confirmation; the nearly $3 million is also a "nearly" figure, and the report does not disclose the precise amount, grant number, or disbursement method.

The project will use Stretch 4, Hello Robot's newest model launched in May. Stretch itself is a mobile manipulator that combines a compact mobile base, a lightweight telescoping arm, artificial intelligence, and a user interface for operating in everyday environments. This positioning determines its research path: it is not a specialized device fixed at some workstation, but something meant to move, reach, and interact in the everyday spaces of homes and care facilities.

The project is titled "Stretching their Reach: Robotic Support for Older Individuals with Mild Cognitive Impairment and Early Alzheimer's Dementia." The title itself makes the goal clear—extending the robot's reach into the lives of older adults with cognitive impairment.

Initial testing will take place at the university's McKechnie Family LIFE Home, a research facility designed to evaluate technology in a home environment. Testing first in a simulated home environment and then moving into real homes and care facilities is the sequence of this research path.

Researchers will apply advances in AI-driven language, planning, perception, and voice interaction in three areas. The first is cognitive assistance, including reminders, activity guidance, and help finding everyday items; the second is social engagement, including initiating interaction and helping maintain connections; the third is safety management, including support for mobility, drinking water, and daily routines.

These three tracks cover the areas where things most easily go wrong in the daily lives of older adults with cognitive impairment: forgetting things, failing to connect with people, and struggling with basic self-care. Concentrating advances in language, planning, perception, and voice interaction on these areas means the project's technical assumption is that AI progress is already sufficient for a robot to understand and intervene in these high-frequency, low-complexity but high-value tasks.

However, the report does not explain the specific technical implementation details or maturity of Stretch in these three areas. Which capabilities are already usable and which are still early is something outsiders cannot judge from the available information.

The project will conduct a participatory design study involving 60 older adults, care partners, and professional caregivers. This is not about having subjects try a finished product at the end, but about putting these three groups into the design process itself.

Home evaluations will involve older adults with mild cognitive impairment or mild dementia and their care partners, while professional caregivers will conduct an eight-week Stretch evaluation in care facilities. The home setting and the facility setting are handled separately, and the evaluation subjects and durations differ: the home side is a combination of older adults plus care partners, while the facility side is eight weeks of continuous use by caregivers.

Hello Robot notes that researchers will evaluate Stretch's effectiveness, usability, reliability, trust, acceptance, and ability to operate safely around older adults. Among these six dimensions, trust and acceptance deserve particular attention—they are not purely technical metrics, but subjective judgments about whether older adults and caregivers are willing to let this machine stay by their side.

Hello Robot CEO and co-founder Aaron Edsinger and University of Illinois professor Wendy A. Rogers are the project's principal investigators, while Hello Robot CTO and co-founder Charlie Kemp and University of Illinois researchers Raksha Mudar and Harshal Mahajan are co-investigators. Two company co-founders plus three university researchers make up the project's core team.

Edsinger said: "The value of a robot is not in how complex the task is, but in how much autonomy that task gives a person," and said the research will help understand how Stretch can provide useful and trustworthy assistance while supporting family members and professional caregivers. This statement lays out the project's value judgment plainly: the measure is not how difficult a task the robot can complete, but how much room it leaves for older adults to keep doing things themselves.

Hello Robot notes that this work builds on prior NIH-funded research by Hello Robot and the University of Illinois. In that phase, researchers improved Stretch's capabilities for older adults with cognitive and physical impairments and evaluated the robot in the natural environment of the ClarkLindsey Life Plan Community, and Lindsey Communities will continue as a research partner.

This continuity shows that the new grant is not an isolated attempt, but an expansion built on existing collaboration and existing scenario validation. The prior work evaluated Stretch in a natural environment, while the new project expands the scope to home evaluations and an eight-week care-facility evaluation, and raises the scale of participatory design to 60 people.

This funding will help prepare Stretch for use in homes and care facilities. In other words, the goal at the current stage is to "get ready," not to declare that it is already usable.

All of the above information comes from the same report, with no other independent source cross-verifying it. This means that from the funding amount to the project timeline, and from the 60-person participatory design to the eight-week evaluation arrangement, there is currently only this one information outlet.

The specific points of uncertainty include: the project start date of September 22 and the end time of May 2029 both come from the company's account, and the report provides no independent confirmation beyond the year; the nearly $3 million is a "nearly" figure; the claim that Stretch 4 launched in May does not specify the year; and the report does not state the grant's specific number, funding mechanism, or disbursement method.

More critically, the report provides no completed test results, performance data, or third-party evaluation conclusions. Therefore, as to what degree Stretch can achieve in cognitive assistance, social engagement, and safety management, the available information can only support "the project plans to study these directions," not "it has already been proven effective."