One of my greatest fears is doing something ridiculous in my house late at night, because you know how I like to shuffle around and I hurt myself and no one hears me call for help. You know, No, that is scary, right it is. But what if instead of laying on the ground needing life alert, there was a robot that could catch me before I even feel I love that idea.
So today we're gonna be talking elder care engineering and the soft, squishy future of robotic support. I'm TT and I'm Zakiah and this is Dope Labs. Welcome to Dope Labs, a weekly podcast that mixes hardcore science with pop culture and a healthy dose of friendship. In today's lab, we're talking about aging in America and the technology that could help us stay safer for longer. Our guest is Roberto Bowley, a mechanical engineering PhDc student MIT who's designing a robot
that can literally catch you when you fall. It's called Ebar. But before we meet the robot, let's set the stage.
What do we know? So between twenty ten and twenty twenty, the US Census told us that the over sixty five population saw the largest and fastest growth spurt that has seen since to like late eighteen hundreds, and according to the Urban Institute, by twenty forty, we're expecting one in five Americans to be over sixty five. And what do we want to know?
Well, I want to know what kind of tech could actually help? Like, how do you build a robot that doesn't feel scary, you know, not like the ones that from I robot that are going to turn on you?
And yeah, what were they thinking? Or that's too invasive? You know what I mean?
Yeah, what does it take to go from a great idea to a real life idea that's manifesting in front of you, that is safety certified and in someone's living room.
I think we're ready to jump right into the dissection.
My name is Roberto Bolly. I'm a graduate student at MIT studying mechanical engineering.
Where do you think some of our biggest disconnects are between people's hopes for what that kind of aging looks like and what our current caregiving support infrastructure delivers today.
So we actually we did a lot of interviews with elderly people where do they need support? What kind of support are they looking for? And we found like virtually everybody we interviewed once to Asian place at home, Like that's like, I feel like the goal. It's like, you get older, you live at home, you have hobbies like gardening or moving around the home, and you know, you just gracefully age in place. The reality is that something like thirty or forty percent of elderly people fall each year.
Oh and oftentimes when they fall it's like a debilitating injury. Like, if you're young, you fall, it's not to make a deal. You might get a bruise or something. But if you're old and you fall, you can sometimes break a hip. And so unfortunately, what we see is a lot of people get shuttled into nursing care or long term care, and they often don't like it. They want to go back to their homes. Sometimes they get suboptimal care because there's also a big shortage of caregivers.
Can you clarify what age is elderly?
Ah?
I think in our lab we've been looking at people over sixty five, but there's no really good definition. Because I met someone who is seventy five and who bikes like eight miles a day is in probably better shape than me, but more than like an age we're looking
at like a subgroup. So we say people who have like medium muscle strength, so they're able to hold onto handlebars or do activities of daily living, but they may lose their balance or they have a tendency to fall, and sometimes for hard transfers like getting out of a bathtub, they require assistance, like they need to grab onto a handlebar or something.
When we consider the increasing demand for care of an aging population, you and your team are proposing robotics as a solution or at least an enhancement. And from what you've said, people want robotics, which I'm surprised by. I think I would have intuitively thought that older people would be anti robotic.
I would have thought that too, But then I mean, here comes Ebar, this really amazing invention of yours that's meant to help older people in the ways that they want to be helped.
Yeah, for sure. So Ebar came about because we're looking around through the literature and through what devices are available, and we found that for robots that actually catch a fall, pretty much all of them you have to wear a harness or like some sort of wearable device. But like elderly people hate to do that. The feedback we got
was that it makes them feel old and it's sometimes cumbersome. Yeah. Yeah, So we're trying to develop a robot that can catch a fall without any like having to wear any sort of device. And then another thing we found is that a lot of elder care robut it's only a few that have looked at physical assistance. But typically you have to stand within what's called the base of support of
the robot. So if you imagine all the points of the robot that touch the floor, if you like connect them with lines, as long as you stand within that area, the robot will never tip because you're within the base of support. But if you're outside, for some robots, they can tip over. So we're trying to design a robot that stable even if you're outside of that base of support, because then it can go like over bathtuble lips, it can go like onto a bed, it can go over
like gaps and obstacles. So we put these two together and we developed what my professor calls like a mobile forklift or robotic candlebars.
I think that that is so fascinating and such a leap forward in the technology for the folks that you know they're going to be listening. They can't see us. Hopefully they'll do their googles to be able to see ebar. Can you just describe what ebar looks like from top to box so that somebody can visualize it in their mind.
Sure, yeah, I'll try to do my best. It's this robot that can drive around. It can catch people with airbags when they fall. It can like physically lift them up with this sort of U shaped fork at the front, and then it has handlebars for them to grab onto. It's kind of just this big U that's padded with handlebars on the front and on the back. That's the part of the robot that you grab onto. You can rest your forearms on it, or you can just grab onto the handlebars on the side. And it has the
airbags underneath which can grab onto your waist. So that U shaped fork is attached to a big linkage, and then that whole thing is attached to an omnidirectional drive base. And so the drive base, you know, like if you're trying to parallel parker car, you can't just like slide into a spot. You kind of have to move back and forth.
Right, Oh, Yeah.
So this robot has four wheels. Each wheel can independently rotate and translate, and so you can actually just go and then move sideways instantly, so it can move in any direction.
That's great, That is amazing.
I didn't put this in the paper, but I believe it's probably the world's fastest elder care robot. So we limit the power for safety, but it has a max speed of around twelve feet per second. It's very powerful, it's moved.
I need one of those for my home.
We limit it to like one to two feet per second. You know, they don't want to cause any injuries.
Yeah. I think that's a great description of the design of how ebar looks. And I watched some of the footage that you all share it's on YouTube, and it looked like you had I don't know if it was PlayStation or Xbox style, like a joystick on there, a controller. Was that part of early testing? You know, what kind of control system does e bar have?
Yeah? So right now, in the original paper, it's just manually controlled with a joystick, like similar to what you see with mobility scooters, where there's a joystick and people just sort of move it around. Currently, right now in the lab, we're working on automating some of the functionality. So we have a project where we're trying to use the drive base to track a person automatically so it
can follow them around. And so you know, it's a lot of work to do that sort of automation, so we kind of just went for manual control first, but we're definitely looking into it.
That's really cool.
My background is also engineering, and so I also have a degree in mechanical engineering, and so I know that part of this is about footprint, because if you're thinking about the function of this robotis to help folks with getting around and to keep them safe, we know that it has to be able to take up as least
amount of space as possible. Can you talk about the role that the robots footprint played in your design and how did you make sure that it could navigate in tight spaces inside of a home without tipping over or interfering with other things in the house or the user.
Yeah, for sure, for sure. So our goal was to make the robe but as small as possible. The problem is that you know, if it were just like six inches wide, it would tip over immediately. So what we did is. We set up a sort of an optimization problem, and from a high level we said, okay, like how small can we make the drive base and how heavy can we make it so that it doesn't fall through the floor. Because you know, if you think about a person, a person occupies about one square foot of space if
they're standing up straight. He said, okay, you know the average person. A floor can support a person that weighs maybe two hundred three hundred pounds, no problem, So we'll try to limit the robot weight to that. Then we looked at like what are the maximum forces people can apply laterally and horizontally. So it turns out the US military has done a lot of studies like this in the eighties and they found that, like I think it's like one hundred and twenty Newton's horizontal force, they have
like force from any orientation. So we can quantify this really well, like when you're standing up, how much can you push on the robot?
Right?
So we put all that together and we have a cost function. With the cost function says, okay, how can I shrink the robot as much as possible while still satisfying the load bearing constraints? And so putting those all together, the actual wheels of the drive base occupy around a ten inch square and then we have two sort of outriggers like almost antennae, so we can get a little bit more stability. And so the footprint of the robot, I think it's around the base itself is around fifteen
inches by fifteen inches, not including the outriggers. Wow, with the outriggers, it's a little bit bigger. I think they're around twenty inches wide. But your average doorways are somewhere between twenty eight to thirty six inches, so we still got plenty of space around the door.
That's amazing.
And you know, I know that this is designed for elderly folks that need help around the house, but I can already imagine this helping a lot of people who are differently abled from the disabled community. And did you consider that as you were designing? Is that something that also came into your mind as another use for ebar?
Sort of, So we did think about using it for Parkinson's patients because they sometimes just have a struggle maintaining their balance, But I think absolutely it could be applied towards different populations all be honest, we specifically designed it for elderly persons, but I think you know, there are certainly other people who could benefit from the robot.
Absolutely. I think that's such a good point that TT's talked about before in engineering and design. We've talked about it in just inclusivity across the board in anything that you create. You know, the more you think about who this can help, the more people benefit from it, even people you didn't consider. I want to know about how you're these different types of assistance that ebar provides, because
it's not just that you have to fall in. Because of the shape of it and because of how you designed it, it seems like it can do more than just like wait until you're at the ground and needing to be lifted, right, like ebar can come in ahead of time. Yeah, I'd love to hear you talk a little bit more about the different falls you anticipate it, or the different types of knees you anticipated, and how you build all those things together into this system.
Yeah. For sure. Our goal was to catch a person before they fell, because when they're on the ground, oftentimes they may be unconscious, they may have passed out, and it's kind of a bad situation to be and if the person's already on the ground, right, So you're thinking of how to do it, and we came up with the idea of using airbags because they're soft, you can have like a large contact area. And the question is,
like is it physically possible. So a previous student in my lab has worked on like fall prediction and she found that you can predict a fall up to two hundred fifty milliseconds before the person actually starts to fall down. Wow, And that's why using like a waste mounted sensor called an im you. So we said, okay, two fifty milliseconds,
that's our target inflation time. Can we like fully inflate the airbags and catch a person right before they so then, yeah, we tried different shapes and sizes and configurations of airbags. I tested a lot on myself, you see, like how how much can I inflate them before it becomes painful?
And then we looked at studies of like skin bruising because you know, we know that elderly people's skin are kind of sensitive, so we didn't want to cause them any danger or put them in risk of bruising their skin. And we settled with a configuration of four airbacks. There's two big ones on the side and then two smaller ones on the front like columns, so when they inflate, they sort of push the person back into the robot.
And then we developed a rapid like two stage inflation system that can inflate them within two hundred and fifty milliseconds, so if you're walking, you know, right now. Again, it's all done manually, but we have done work in our lab about like detecting falls, just haven't implemented it on
the robot yet. The idea is, if you begin to fall, the airbags which just rapidly inflate and gravy and sort of hold on to you, and so you're kind of frozen in this position, but you have a chance to regain your balance, or we can deflate the airbags and you can continue normally, or we can just hold you, you know, until someone can come to help.
And so the next part of the engineering is to you know, test it out on the users. I'm wondering, you know, one, I think it'll be interesting to hear like how you get volunteers for this or who you recruit to test these things out, and then also what their feedback was and how that informs like how you move forward and any changes that you might make.
The major thing we have to make sure before we can roll actual elderly persons that it's completely safe because we don't want to cause anyone any injury. So right now we've been testing in healthy volunteers. By healthy volunteers, I mean myself and a couple of my web mates. Okay, but yeah, the idea is to refine the system to make sure it's safe, to you know, sort of measure the forces which we have been and so far everything's good.
Then we can apply for approval from our university and then start enrolding elderly persons.
Okay, this sounds really cool because you've called EBAR a step towards aging in place. And I am thinking back to those commercials where people were like life alert and they were waiting until they already fail. And like you said, after you've once you hit the ground, anything could happen. You could be unconscious. But like I think about what this could mean. Something you mentioned is how maybe stressed
our healthcare system is for aging adults. Yes, we're talking about this in the home, but could you imagine this type of tech in other places like care facilities or as part of public programs. How do you imagine this tech? You know, I know we're looking much further ahead, but how do you imagine it scaling?
I think for me personally, i'd say, like, I think the gold standard is a human being. It's very difficult for robot to replace a human, right. I think We've seen this again and again. But the reality is that if there is a care shortage, I think it's better to like augment the shortage with robots than just to not be able to do anything about it. So one thing we were thinking is like in nursing homes, ebar
could sort of handle some of the easier tasks. If a person just needs to walk, for example, to a sink, then ebar could help them out. But very complex tasks like lifting a person into a bathtub with a sling like those can still be done by the caretakers, so it can sort of free them up. That instead of elderly persons having to wait a long time for care that if for some tasks we could send the robot and for some tasks we could send a human caretaker.
I love that because Menzakia talk about this all the time where a lot of solutions to the world's problems is not an or response like this or that. It's an and so not to say, oh, Ebar is going to replace all caregivers, there won't be a need for a human. It's like, no, you can have the human and you can have e bar, which to supplement and it makes it an even better experience. I'm curious about,
you know, what are the next steps with ebar. Are there any like upgrades or advancements that you want to add to EBAR, or what's the vision or are there things that you want to do outside of e bar to also help supplement EBAR.
Yeah, that's a great question. Honestly, it's discussion my advisor had with me almost immediately after I submitted the paper. Yeah. I think certainly there's room for making the robots smaller and more compact. I've talked with a lot of people, even my parents are saying like, hey, you should add something that can pick up a person's phone if it falls down, like a sort of a coup holder on the robot. I said, well, that's great. I'd love to do that, but I don't know if I can really
put that in my thesis work. So one of the other things our lab is looked into is handlebar optimization. Like if you think about it, when you go into restrooms or some public places, you see handlebars on the walls. It turns out there's not really been a lot of work that's been done on like is this the biomechanically optimal location for handlebar? And because like you know, who knows, it could be in front of you, it could be
like up here, it could be like down there. So we were trying to look into, like can we make a model of a person and then predict like where is the optimal place for handlebar it maybe reduces the muscle strength the most or provides the most support. So that sort of stuff I think is interesting and I think could compliment a robot like e bar because then it helps us nowhere to position the U shaped fork
based on the posture of the person. So yeah, that's I guess that sort of work that we've been looking to and thinking about pursuing in the future.
Okay, I think this is so cool.
Things.
You know, this is totally outside my comfort zone of science. TT understands this kind of stuff way better than me, which is why she has such great questions, is there anything else here seeing in the field of robotics and elder care? Is there anything else you're excited about this in the research pipeline, even if it's not in your lab, like with other labs that do great work that you see, Ye, that feels promising.
So we are collaborating with So just full disclosure. But at Stanford, Alison O Kumurro, she did a lot of work on these things called vine robots. They're sort of like flexible tubes. They can inflate and they can go under a person. And so we have this idea. It's like, wait a second, you have an elderly person lying in bed and you want to like put them in a sling, Like right now, you've got to lift them up and
pull a slip. What if the sling just like inflated under them, like and then you could just pick them up that way. So that work I think we've been doing with her. But she was the one who's been working on vine robots. I think that's really cool. In terms of other labs and other research, we see a lot of work with humanoids. Humanoid robot seems to be
like the next big thing. There's a bunch of companies pursuing humanoids, and so people have started looking into like using a humanoid for sit to stand assistance, which I think is pretty cool. My concern with it is just, especially if the robot has two feet, it's not going to be a stable as something that has like a heavy, big drive base. You know, like if the robot and the person fall down, then that's really bad. But I think it's great work though. I think that that should
be explored. And I saw a couple of papers when I was at akra Acra is a big robotics conference that happens each year. It's sort of like v big robotics conference, and so I saw a couple of papers where people are looking at using humanoids to like provide sit to stand assistance, and I thought that was really cool.
Yeah, I was thinking about this the other day because I've also been noticing a lot of the humanoid innovation and how it's like, you know, as engineers, we try and create things that are not just flashy but useful, and it doesn't really seem intuitive to me to say, this human is not able to form this task, so I'm going to replace it with another human but I think people are just so enamored by humanoids.
I'm not. As a non engineering person, I feel like, give me the e bar. If I knew I was falling back into a face, I don't think I would like this, you know, And so I think it's interesting, Like I feel like that humanoid component is what makes people feel like, oh, you're trying to replace me, or oh that's my own bias that I think. If it's hard already for a person to help you, why would I put it in something else that's person shaped to do the same work so it can be just as hard for it.
Yeah, make that inflatable sling, Like, I love that idea.
I think it's great.
My last question is just about the process, because we just did an episode on the research funding cuts and stuff like that. I really would love for you to highlight like what it takes one to get to this point with EBAR and then to have EBAR kind of like break into the healthcare system, if you know, because I think that one thing that the Kia said is that folks they don't think about research in the way
that they should. They think about innovation. So when the product is there and in their hand, they're like, ah, research, and it's like, no, that's innovation. There's a lot of research that gets you to that point. So can you talk about that process for yourself. I think it would be great for people to here to understand what it takes to do something like this.
I took product design. It was a class at MIT a couple of years ago. They presented us with this process. I think it's just called like a need driven develop and process. But the idea is you start first with the stakeholders. You go to the elderly person, you ask them like what sort of tasks are you trying to do? And then oftentimes what you find is people will tell you things, but really there's like underlying needs they're called
latent needs that they're not really telling you. They can't really put into words well, but it becomes obvious as you look into more of the of what they're doing. Like, for example, an elderly person might say like, oh, you know, I really have trouble getting into and out of the bathtub, but I want is just like like a cane or something that'll help me get in and out. And you're like, well, maybe the problem is that the tub is too high.
Maybe it's the problem that, like, you just need a little bit of support, and so it may not be a cane that you want. It may be like a robot or a handlebar or some other device that can also help you. So we start with this stakeholders and looking at the latent needs and then in our case, we develop this sign concept for robot and we went
to a physical therapist that's Spaulding Rehabilitatian Hospital. We did a presentation and they said, oh, we like these aspects of it, but we think these aspects wouldn't work with the patients that we work with. So then we sort of go back and forth a couple of times, rEFInd the design, and then from there on it's the most fun part for me. I get to build a prototype.
I think that's interesting because that's such a long process and the only research you're highlighting is the new research
that you're doing. But you've already told us that even in your prototype, to determine how much force a person might use, you're going back to research from forty years ago to the nineteen eighties, right, and so research that some people may say, why does the army need to do this, or why does the military need to figure out how much force a person exerts laterally, because in forty years, a graduate student is going to be figuring out how to design a robot that can keep your
grandmother from falling when she reaches for something. And I just think those are the connections between research and innovation that we don't highlight enough. You're relying on research that has already been done to help you iterate and innovate really quickly to decide what the drive based size should be. And I think it's just such a good demonstration of research and innovation how closely those two rely on each other.
For sure, for sure, And there are so many times when, like, like, one of the critical things that we've needed was you needed to know the friction between a person's skin and their clothes. There's a sony in twenty thirteen that characterizes so thank you so much. You know, it's such an obscure piece of knowledge. You think, like, when would you ever need that? But it turns out we need to know, like, well, the closes slip on the skin if we can press
them with airbags. So this random research study ended up being extremely useful because we can quantify now the friction. So yeah, there's a ton of these connections between prior research, some of it not even remotely related to elder care. But you know, it's like we're standing on the show of giants. Yes, all this knowledge gets built up, and you may not think it's useful in the moment, but then twenty years down the road someone needs it. It's like the exact thing that they need exactly.
And that's another thing that we always say is that science never stops. And that is a really good illustration of that, where you know, just because that part of research ended or that paper was published, doesn't mean that it'll never be used again. It doesn't mean that it won't be useful. It doesn't mean that you can't build
off of that. And so like science never never stops, someone will be citing your paper one hundred years from now, I'm sure and saying, wow, look what we were able to do based on what Roberto and his lab group were able to do, and now we can innovate off of that. The science never stops, and the work that we do is really important, even if it feels obscure. Sometimes I think about my graduate, my dissertation work, and I'm like, oh, oh god, no one needs this. It
was in nano materials. And then I'll go and look at some of my old papers. I'm like, wow, cited eighty times. Okay, that's not bad, that's excellent.
Yeah. No, it's cool to see you. Everything fits together, and honestly, I think we're going to see more and more on that.
Congratulations on all of your success. We will be tapping into mix to see how Ebar is evolving and your success as your career continues to grow and change. You're doing a great job and doing a lot of really important work. So congratulations, Well, thank you so much.
I'm honestly, I'm just so grateful to have the opportunity, you know, to be a graduate student and to study all of this. I'm you know, I'm always very grateful for that.
So amazing.
Thank you, guys. I appreciate it.
You can find us on X and Instagram at Dope Labs podcast.
He is on X and Instagram at d R Underscore t Sho.
And you can find Zakiya at z said So.
Dope Labs is a production of Lamanada Media.
Our senior supervising producer is Kristin Lapour. And Our Associate producer is Issara Savez.
Dope Labs is sound design, edited and mixed by James Barber. Lamanada Media's Vice President of Partnerships and Production is Jackie Danziger. Executive producer from iHeart Podcast is Katrina Norvil. Marketing lead is Alison Kanter.
Original music composed and produced by Taka Yasuzawa and Alex sugi Ura, with additional music by Elijah Harvey. Dope Labs is executive produced by us T T Show Dia and Kia Wattley.