Creating a successful robot company based around providing commercial services is not easy, although as of just the last few years, advances in robotics technology has at least made it possible. Companies like Savioke have shown that robotics has reached a point where autonomous platforms can operate in semi-structured environments, doing useful tasks reliably and cost effectively enough to make a compelling business case.
Luvozo, a startup founded in 2013 and based in College Park, Md., is bringing autonomous robots to semi-structured environments with an enormous amount of potential: skilled nursing facilities for seniors. They’re introducing a “robot concierge” called SAM, designed to “provide frequent check-ins and non-medical care for residents in long-term care settings” through autonomous navigation, telepresence, and an innovative fall hazard detection system. The potential market here is enormous, and to find out more, we stopped by Luvozo and spoke with CEO and co-founder David Pietrocola.
Here’s an overview of SAM; note that the physical robot you see in the video is an earlier prototype that Luvozo used for pilot projects, while the rendering (the slick-looking white version) is how the robot looks now:
SAM’s main function is as an autonomous telepresence platform. One robot is designed to cover one floor of a senior living community, which works out to about 25 residents and two to three aides. In many nursing facilities, staff spends a substantial amount of their time checking in on residents, once every hour or so (more frequently if, for example, the resident is a high fall risk). SAM takes over this task by autonomously navigating from room to room on a check-in schedule. Once at a room, remote staff can use the robot to videoconference with the resident in that room, making sure that everything is fine. If there are any problems, the remote staff can contact local employees to get things sorted out in person.
There are lots of advantages to using a telepresence robot in this situation, and none of them involve replacing human workers: it’s all about giving residents the attention that they need. The robot takes care of moving from room to room all by itself, which means that remote staff can be talking to residents continuously by hopping into different robots on different floors, or even at different facilities. Local staff, who are already very busy doing all kinds of other things, don’t have to check on residents as often, freeing them up for other tasks. Meanwhile, the residents themselves are getting a lot more interaction, albeit of the robot-mediated sort. The upshot is that local staff are less restricted by check-in schedules and can spend more time focusing on the residents that they do interact with in person, without reducing staff availability overall.
It seems like there might be some concern about whether or not residents accept having a telepresence check-in as opposed to an in-person check-in, but remote staff have access to a detailed social profile on each resident that they talk to, so every interaction is, effectively personal. And in pilot studies, David Pietrocola tells us, residents seem fine with it:
“In our pilot, in most cases, it wasn’t an issue. From the resident’s perspective, it’s about a more personalized experience. During the pilot, we’d have SAM just sitting there in the hallway, and by the end of the pilot, people would come by and say, ‘Hi SAM!’ even though there’s no response mechanism built in. It made the system part of the community, in a way, and the fact that we’re using robots as opposed to tablets or something adds a different dynamic. We don’t really equate a SAM check-in with an actual person checking in. It’s more about serving as a channel by which residents are getting additional attention.”
Luvozo says SAM (including the service that comes with SAM) costs approximately one quarter of what a certified nursing assistant does. This is a substantial amount of cost savings, but what’s also important is to consider what’s happening in this space in the near future. Populations are aging quickly, and there simply aren’t enough skilled workers willing or able to take the jobs that are available: Over the next 15 years, the caregiving workforce will need to triple, and there are already tremendous issues with turnover. SAM acts as a force multiplier, allowing the same number of people to do more, and provide better care. Furthermore, SAM allows workers who are already in the caregiving industry to keep working longer, on their own schedules and from home. It also makes the job location independent: a remote worker could live wherever they want, and as long as they have a good internet connection, check-in on residents in facilities around the world.
With such a clear value proposition here, we asked Pietrocola and Luvozo chief strategy officer Michael Austin why there aren’t already a bunch of other companies in this space. “Navigation has come a long way in the last couple years,” Pietrocola explains. “The fact that ROS has matured so quickly and so well is one reason why something like this can be deployed. And part of it is that nobody has put the right product set together with the right business model to make it work.” Austin adds that “there have been some attempts to put a robot like this into a home environment, but that’s just so varied, every home is going to be different. So the scope of our system is based around an environment with a little more conformity that’s easier to navigate.”
As with any robot designed to autonomously navigate around people, SAM has plenty of features to keep it safe. The robot weighs around 41 kilograms (90 pounds), with a custom suspension supporting a low center of gravity that makes it very stable. A four zone bumper system (which, ideally, should never be used) backs up IR sensors all around the perimeter of the base. The main navigation sensor is a new 10-meter SICK lidar (with ⅓ of a degree angular resolution), which Luvozo is helping to test out in a robotics application, and there’s also a RGB-D sensor used for additional navigation data and for SAM’s fall assessment system, which we’ll get to in a bit. The final piece to SAM is the human-robot interaction component: a 19-inch screen, a webcam and microphone array, and some nice speakers powered by an amplifier (important when you’re spending all day talking to people who may not have the greatest hearing).
Besides the autonomous navigation, one of the main features of SAM (which differentiates it from a pure telepresence robot) is Luvozo’s AFAS, or Automatic Fall Assessment System. Using the RGB-D sensor (a Kinect 2, at the moment), SAM is constantly scanning the floor around it for environmental fall hazards: things like clutter and poor spacing that could cause people to trip. This is a bigger problem than it may sound like, especially because the consequences of a fall are often so severe, and many falls are preventable. AFAS can take measurements and then alert staff with specific recommendations on how to fix problems. This capability is, in fact, one of the primary reasons that care facilities are interested in using SAM.
Last fall, Luvozo ran a pilot deployment with SAM at a Washington, D.C. senior living community. “After that experience,” Pietrocola tells us, “they came back and said, ‘we want this for real, full time here.’ ” This is the best way I can think of to end a pilot, and Luvozo is currently working to close an investment round and finish product development so that they can start delivering SAM to their initial paying customers by the end of the year.
Looking forward, there’s a lot of potential for SAM to expand its role in the managed care setting. Conversational AI is one option that Luvozo is considering to give SAM more autonomy, but they’re also exploring the idea of a family portal, where SAM can be used as a more traditional two-way telepresence platform that allows residents and family members to videochat with each other on demand. In the short term, though, Luvozo is focused on addressing the needs of their direct customers: the management and staff of retirement communities, and the residents that they care for.
[ Luvozo ]
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.