Glidance: It's not a cane. It's not a dog. It's a self-driving mobility aid.

For years, technologists have experimented with ways to assemble powerful new technologies like computer vision, digital navigation, and a variety of  sensors to help blind and visually impaired people navigate more easily. Former Microsoft engineer Amos Miller, who is blind himself, had an idea: why not create a device that uses multi-modal AI technology to guide users by attaching the familiar concept of a cane to a small, two-wheel assembly that guides with steering and brakes a user to their destination? Could people, especially those who lose their vision later in life, easily afford the device and use it right out of the box? That’s what Miller aims to deliver with Glidance. Immediately after this session, Amos Miller will be available for live questions in a breakout session listed in the agenda.

[MUSIC PLAYING]

JOSH MIELE: Thanks so much, Alice. It is such a pleasure to be here at Sight Tech Global ’23. And today I’m speaking with Amos Miller, who is a well-known innovator in the area of wayfinding and orientation mobility technologies that support blind people in independent travel. And Amos, always great to talk to you. But for those who aren’t already familiar with your work, do you mind introducing us to a little bit of your background?

AMOS MILLER: Josh, thank you very much. It’s really good to talk to you and to be here at Sight Tech Global. I think love the energy of the innovation around this event. I’m the Founder and CEO of Glidance which we’ll talk about in a few minutes. But prior to that, I spent many years at Microsoft Research. Most recently, I led the development of Microsoft Soundscape which many in the community will be familiar with. This is technology that uses audio augmented reality to help a person build a richer awareness of their surroundings and navigate effectively.

JOSH MIELE: Yeah, and one of the cool things about Soundscape, I’ll just mention, I think is its use of spatial sound, right? You mentioned auditory augmented reality. But basically, what that means for people who aren’t in the tech world is you can hear where things are just from a pair of headphones. It’s amazing.

AMOS MILLER: That’s right. And the goal there is to enhance awareness. So you walk around with your dog, or cane, or residual vision, or sighted guide. And it enhances your awareness. And I would want to add that I am blind myself. I lost my sight in my 20s from [? RP. ?] And so technology and sight loss is a passion of mine and has been for a while. And I continue to explore ways in which advancements in technology can really make a difference in our ability to get around and live life to the fullest.

JOSH MIELE: Absolutely. And yeah, we share that. And I’m sure many of our folks in the audience share that as well. So, Amos, just for those who don’t necessarily use the tools that we use to get around, can you give folks a quick overview and orientation to the landscape of the tools that we use– the methods that we use as blind people to get around in the world?

AMOS MILLER: Sure.

JOSH MIELE: Yeah. Thanks.

AMOS MILLER: Yeah. No, that’s a great question. And thank you for bringing it up. And I’ll give my view here. It’s not necessarily everybody’s view. But I think that really in the last 15, 20 years, there has been incredible innovation in the mobility space with a variety of navigation apps that are getting better all the time and the equipment that they run on the phones, smart canes, tactile maps, now more live technologies. One of the things that really blew me away very recently, Josh, if you played with the generative AI that provides scene description. And you can put up your camera and really get a layout of the space and navigate through that. So I think we are blessed with incredible innovations.

JOSH MIELE: Yeah.

AMOS MILLER: But I think at the same time, it’s very important to acknowledge that the benefit of these technologies are mostly limited to people who are already out there and confidently use their canes or guide dogs. And one of the things that I learned over the years is many people in the community are not as confident. Many have not necessarily received a great deal of training. And they don’t get around independently. I spoke with a student not a few weeks ago that told me about her. She studies law in a reputable university. And she has to choose her classes based on the routes that she has to walk rather than what she wants to take.

JOSH MIELE: Yeah, so I mean, the tools the choices that we have often do shape the decisions we make about what opportunities we’re able to connect with. And you mentioned all these innovative things. And I just want to underscore that at the bottom of it all is the low tech independent travel, cane skills, guide dog skills, knowing how to walk with a human guide. And so all of these other technologies tend to be additions to those fundamental skills. But you’re actually working on something really new. It’s not an app. It’s not a tactile map. It’s not part of a smartphone. It’s a departure I think from many of the other technologies that we already have to augment the tools that we’re using. I think it’s called guidance. And you’ve got a whole organization that’s around it. Could you tell us about the mission of the organization and the project that it’s focused on?

AMOS MILLER: Thank you, Josh. Absolutely. And really to point that out, the challenge that we’re trying to focus on is really how do we help many, many more people to get out and about. And that’s where I found the Glidance. Glidance is a new company that focus on to help advance the field of independent mobility with AI and robotics. We have a world class team. I’m very fortunate with collaborations with a number of universities. And our mission really is that every single individual with sight loss can get around independently. And I know that’s very ambitious. That includes people who lose their sight later in life who many of them struggle to develop strong mental mapping skills. And to do that, we believe that there’s a need for a new primary mobility aid that we call a glide.

JOSH MIELE: I think you’ve got a video that gives a quick demonstration of the concept. And I think let’s play that now and then we can talk about it more.

AMOS MILLER: Wonderful. Let’s do that and I’ll explain afterwards how it works and everything. [VIDEO PLAYBACK]

AMOS MILLER: Hi. I’m Amos Miller, and I’m the CEO of Glidance. And we’re introducing glide. A new mobility aid for people with sight loss that uses autonomous guiding technology to simply guide the way. And I’m going to show you how it works. OK, I’m standing in a shopping center. I have glides handle in my right hand. And the handle extends to about two feet in front of my right leg. And it rests on its two wheels. Now as I nudge it forward, the wheels start to steer the way. SPEAKER 2: Using its smart sensors and camera, glide spots a pillar in the way and is steering its wheels to guide Amos safely around it. While walking through a crowded area, glide weaves a smooth path. Amos is walking confidently and with purpose. Take note that glides wheels are not motorized. Glide is not pulling him at all. He’s walking at his own speed. Amos then twists the handle to turn left. And glide takes the turn to the left only when it is safe to do so. With its sight sensors, glide helps Amos keep a nice straight line along a corridor. Now using its camera, glide detects and steers Amos to the escalators. With a haptic signal through the handle, glide indicates for him to slow down before smoothly boarding and coming to a stop. As Amos reaches the top, he nudges glide forward off the escalators and glide leads Amos off to the right. By detecting the door to a store, glides smoothly and purposefully steers Amos inside. Easy to use light and affordable, glide is a mobility aid for the future. Glides, your way to autonomy and independence. [END PLAYBACK]

JOSH MIELE: Well, really great video, Amos. The product itself sounds really exciting. I think what I’d like to dig into now is, can you give us a little bit more detail about how you’re going to make all this magic happen? I’m very interested in the technology.

AMOS MILLER: Absolutely, Josh. It is what it is about. As I said glide is a self-guided mobility aid that simply guides the way. And it uses sensors and AI to do that. So I’ll just quickly explain. So I’ve got glide in my hand right now. I’m actually showing the bottom of glide where there’s two wheels that touch the ground. And on the body of just above the wheels, there are range, there are short range sensors that detect the environment around glide. It can see objects, it can shoreline. And the way that it guides is that you will it forward. And it steers the way. And I’m just moving the wheels left and right just to demonstrate that the wheels turn like the front of a car. And the really cool thing about it is that the wheels are actually free wheeling. They’re not motorized. So they don’t pull you around. You nudge it forward. You walk at your own speed at your own leisure. And glide steers the way keeps you on a safe path and avoids obstacles.

JOSH MIELE: Can you rolling it out?

AMOS MILLER: Let me just show the top of the handle. I’m following to the top of the handle where the person holds the device. And on the top here, we have a camera that can look further afield, help you detect the destination, detect a door, detect something and really plan a path, and then steer you all the way there. So that’s broadly speaking what glide is.

JOSH MIELE: And, Amos, for those who aren’t able to see the object or the video, just the idea the ergonomics of it is you are rolling the glide. The wheels are in front of you by maybe a couple of feet.

AMOS MILLER: A couple of feet. Yeah.

JOSH MIELE: And you’re holding a handle that almost like a cane, right?

AMOS MILLER: Probably a little bit more of a guide dog handle. So you hold it around your waist level with one hand and you walk forward. If it’s in front of you, if you’re holding it in your right hand, and glide will roll a couple of feet in front foot. And as you walk, the wheels steer. So you don’t have to make your immediate navigation decisions or there’s an object where is there a path to get around it. Glide just takes you around it.

JOSH MIELE: And where are all the smarts for this? How does the actual– so I understand that there’s a camera on it, there maybe some other sensors, probably infrared sensors, maybe a LiDAR thing. I don’t know. But who’s doing the thinking? How fast is the reaction time, stuff like that?

AMOS MILLER: Yeah. No. So really, Josh, this device is planned to be a mobility aid, right? It’s not a robot, and it’s not something that’s a super complicated. Our goal is that when you leave the house, you pick it up, and you walk. You don’t set it up and go and do all kinds of crazy things. Yeah. And so to do that, glide it does a lot of the thinking inside the device itself. So it’s self-sufficient, it has compute capabilities, and so on. But we also have a companion app for it. So for settings and for things that you want to do. But you don’t have to open the app every time you go somewhere. And it will also be powered by capabilities in the cloud, especially for the more advanced AI scene understanding capabilities, which will occasionally need. But it’s not going to depend on that. So it’s all runs on the device, very quick reaction, immediate response to everything that’s happening. So this needs to be to work quickly.

JOSH MIELE: And I think that its primary tools like this have different capabilities. And you demonstrated some of them in the video that we played. But there’s the fundamental obstacle avoidance that is something that I think this is intended to do. But there’s also finding doors, finding paths, finding crosswalks. Is there really enough onboard processing power to bring all that off and without a lot of latency?

AMOS MILLER: Yeah, so one of the benefits that we have with glide is that this is not a wearable device. You don’t have to carry any weight. You don’t have to– it runs on wheels. We can put a bigger computer on it. We can put more battery on it than you’d be able to put on you. We can choose our own components. We’re not dependent on somebody else components. Yeah, and so we can really optimize. And there is tremendous progress in these miniaturized compute AI devices that really don’t cost a great deal of money. And really these are the technological advancements that enable us to get to this point and build this device today.

JOSH MIELE: So I’m really interested in talking about who’s going to benefit the most from this, right? We know that there are many experienced cane users and guide dog handlers out there. Who do you expect to really benefit the most from this innovation?

AMOS MILLER: This is a great question, Josh. I mentioned earlier that the large majority of people with sight loss actually lose their sight later in life. And many don’t have strong cane use skills. And so we are working to design glide to give them that simple solution that gets them back on the path to independence as quickly as possible. So I think that one of the benefits of glide is that it’s so easy to use. That you can probably take a walk in your immediate neighborhood with glide within a day of getting it. Yeah. Which is as you would probably agree, a leap from any other primary mobility that we can get access to today. And now that’s not to say that I think cane users, we have lots of feedback from people who are cane users and people who are guide dog users using the device and benefiting from it and finding real value, and certainly in certain scenarios where they just want to get around quickly and move quickly. I think we have to think about it as another tool in the tool set. Yeah. And we really hope that glide will serve a role in bringing O&M to many, many more people really by enabling you to get back on your mobility journey really quickly, but also for existing cane and dog users. I think that there’s benefits there.

JOSH MIELE: And I think you mentioned the cane skills, the guide dog skills to– if you want to use a guide dog, it requires weeks of training, and lots of– and of course, cane skills also require training and building of skills. And I think confidence is also a huge part of what holds people back as they are building these skills. What kind of skill building will glide require? And what do you think it will do in terms of people’s confidence? How can it support not just skill building, but confidence in your ability to travel independently as a newly blind person?

AMOS MILLER: Look, first of all, I don’t want to diminish the importance of orientation and mobility skills. OK, but I think it’s important to understand that glide ultimately provides navigation that reduces your dependency on your orientation and mobility skills. Now, if you know the area, you’ll navigate on your own. But if you don’t, and there’s a path that you need to take on a regular basis. Glide can take you on that path without you having full orientation awareness of every single step that you take. And that’s an important departure from how we think about O&M today. This is going to guide you. You can guide it you can drive, but it is going to guide you. And that’s a leap that we’re as a community are going to have to take forward.

JOSH MIELE: And I think one of the interesting things about this– and I’m interested in your thoughts also. One of the interesting things about this is that we’ve been talking about the idea of devices. Meaning people like you and me who think about how we can innovate. We’ve been thinking about devices like this for a long time. And for so much of our professional careers, this has been a pipe dream, not something that the technology could support. But it seems like one of the things that we need to do every now and again is reevaluate the ideas that we’ve already said, oh, that probably wouldn’t work. And to reevaluate it in the context of the changing landscape of technology and the capabilities of computers, do you want to just talk about that a little bit too?

AMOS MILLER: Yes, I mean, I think that there are a lot of wrecks along the path of mobility aids. But there’s real gold nuggets in those wrecks as well. Sometimes these wrecks are because of the challenges and the readiness of technology, often, these are just in terms of how they fit into our lives as people who are blind and live a daily life, and how it interferes, or supports our experiences. And I think it’s a combination of those two aspects. Like you said, it’s a pipe dream that– it’s obvious, right? We have autonomous vehicles. Why don’t we have robots that guide people? But to do it and to do it right, we have to have the affordable technologies to do it with the powerful technologies to do it with, but also to design it in such a way that the user remains independent, that the user keeps their agency. But it really benefit from the experience. And those are the things that as we evolve the concept of glide have been sent front and center. I don’t call it a guide dog or a robot dog because it doesn’t work, it doesn’t bark, it doesn’t step on your feet. Yeah. It runs on wheels. I don’t call it a smart cane because it doesn’t work anything like a cane. That’s why I call it a glide. It’s a completely different device. It’s a different experience. And I think that this is definitely something that will start opening up that opportunity to get these technologies to people’s hands.

JOSH MIELE: Yeah. And I think that it’s so interesting to me that know that we’re able to– the changing capabilities of computers really do change what’s possible. And the other thing that I just want to emphasize for the audience is that a lot of the guidance aids that I’ve heard about in the past have been designed by sighted people for blind people with maybe a blind person in the design mix. But you are blind and you are extremely experienced, not only in the building tools around orientation mobility, but you are a user of such tools. And this is a point that’s really important for folks to understand that we as blind designers have insights and understandings of the context in which these tools are going to be used. That are very difficult for many other folks who aren’t blind to really to integrate appropriately. So just a little soapbox of my own. You want to comment on that?

AMOS MILLER: No, I wholeheartedly agree. I think you’ve said it. You’ve said it right. It’s the intuitions and the nuance that is hard for people. People asked, why can’t you use an app just to do it? And I say, imagine trying to walk through a door and not hitting the two sides of the door with something, how do you do that, right? Most people don’t think about that question. But we have to deal with that all the time, right?

JOSH MIELE: Exactly.

AMOS MILLER: Yeah. So those are the nuances. Yeah.

JOSH MIELE: Amos, we’re coming to the end of our time together. I just want to– let’s get concrete. When do you expect this technology to be available and how much do you expect it to cost?

AMOS MILLER: Thanks. So our goal with glide is that everyone gets one. And with that, we are looking at a very affordable device. And it’s designed that way. Our goal is for it to cost somewhere between in what it costs to own a cell phone subscription. So it’s not $50 for the device, but it’s also not $3,000 for the device. Something that people can afford in the range of a cell phone. I can’t, at this stage, provide a release date. But what I can say is that we’re very soon going to be opening a pre-order program for glide. And that for small deposit people will be able to secure their place to get one at a discount. And what I would say is I would urge folks to hop onto our website, glidance.io glidance.io. Glidance is written as guidance, but with an L instead of a U. And just register your interest. And we’ll be sure to let when we open the program, and also keep you informed of developments. We have interesting opportunities coming up in the coming months.

JOSH MIELE: Great. Amos, I’m going to be tracking this project with great interest. And I really want to just say thank you for this conversation. And thank you for working on these kinds of important issues to help blind and visually impaired people to have more options for independent and confident travel. Really appreciate–

AMOS MILLER: Thanks, Josh. I really appreciate it. And I want to say that I’m not doing it on my own. There’s a whole village of people who are so motivated and excited. And I invite people to join us on this journey to engage. This is we’re doing it out in the open exactly because it is so important. It’s so important to get right.

JOSH MIELE: Fabulous. Well, I really look forward to finding out how it turns out. Thank you, Amos, for speaking with me. And great pleasure to be here at Sight Tech Global. Back to you, Alice.

[MUSIC PLAYING]