Events Add an event Speakers Talks Collections
 
FOSSASIA Summit 2020
March 19, 2020, Midlothian, TX, USA
FOSSASIA Summit 2020
Request Q&A
FOSSASIA Summit 2020
From the conference
FOSSASIA Summit 2020
Request Q&A
Video
Open Source Robotics Middleware Framework for Healthcare
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Add to favorites
977
I like 0
I dislike 0
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
  • Description
  • Transcript
  • Discussion

About the talk

The healthcare domain has incredibly complex logistical needs around the clock to keep hospitals and various medical facilities running smoothly. As the growth of elderly populations outpace the growth of the workforce, the economies of scale are pushing healthcare providers towards automating as much of the logistics as possible. This includes delivering meals to patients, keeping supplies and medical instruments stocked, securely transporting pharmaceuticals, assisting patients who have difficulties with mobility, and more.Modern and emerging robot platforms can help address each of these needs on an individual basis, but all these technologies need to be seamlessly integrated a single facility, and no single robotics vendor can feasibly affordably offer a complete integrated solution at the scale that is required. As a vendor-neutral open source robotics software company, Open Robotics is working with several Singaporean government agencies and robot vendors to develop an open source framework and open specification to allow heterogeneous robot platforms from any compliant vendor to cohabitate the same facility. This talk will discuss our current progress in this endeavor.

About speaker

Quigley Quigley
Chief Architect at Open Robotics

Morgan Quigley received a PhD in Computer Science from Stanford University in 2012, after which he joined Open Robotics as its Chief Architect. His research interests include collaborative open source software development, scalable distributed embedded systems, and hardware/software co-design.

View the profile
Share

How do you know thanks for having me? Like my colleague. Luca, that just booked for previously. I'm from the open-source robotics foundations. That's open. Robotics weird have an office here in Singapore. As Lucas said, as well as all around the world. Show me to talk about an open sore, sore buttocks. Pain away from work for healthcare. I'll get into exactly what we mean by that. There's lots of applications of Robotics of in healthcare. Now the first thing that you think I might be surgery robots which is great but there's also many other applications of just the logistics and you

know large hospitals are a giant operation happening and just everything needs to move. Everywhere else, Colonel Thompson course I can't talk for a long and Robotics without saying we're hiring. We are hiring for many different roles ranging from robotics engineers to dive off the web to build test, that used to hold. Everything that needs to happen in, and we're always looking for it for great people interested in this. So let me back up a little and talk about motivation of why we're doing this big project is that you were trying to do and how we're going to do it. So these when we

talked about hospitals, do you know here in a general is Singapore in specific, these are very large buildings and in large cities talking like one thousand to two thousand dead hospitals. So there's a ton of people involved in the situation that may be thousands of people at any given time inside the building. All of them needs to be fed taking care of given supplies. They need inputs and outputs all profits appropriately. So, there's a lot of opportunities for automation to help make this is the process used to run more economically as they can scale to the anticipated needs of the future.

The Inn in general, there are many different vendors of robots that sell into the healthcare robotic space. So for any particular size of Supply that you're trying to move around, there might be a few choices but it's very rare for a vendor to be able to offer the full full range of Supply motion sweat vest from just a little, small medicine bottles all the way up to like a giant cartridge giant pallet full of meals for an entire Hospital word. There's there's a wide range of volumes and shapes and sizes and massive that need to be moved around the facility and again we're not even talking

about surgery about say we're talking to Allie about logistics or cleaning your boss maybe you're just anything that's not surgical. So just listen you can kind of block the block that part of the demesne out and and take more of this sort of facility Management Facility processing helping So we're going to Building Systems, we're starting with open source robotics projects that look, he mentioned called Ross, the Robot Operating System, as well as gazebo, which is a stimulation from work, and we're going to use those. Then to build a to hospital scale, simulation in, multi Fleet

Management, and I'll give them to do the next 45 minutes or. So, what we mean in little more specific about that, So, I'd open robotics for the stewards of two, large open source projects, the Robot Operating System or Ross, we've been working on that in public now, for about 13 years or so. The first thing official version had a number on, it may be in 2007 or 2008 and you don't answer to go back quite a bit further than that that the demands are in the lower left Ross. Orgen in there. Say a new and improved, everything better Ross to. Org also happening in parallel. That project

Works hand-in-hand with the stimulation from work called gazebo the the URL there on the lower right there cuz he will send out of work and then the next Generation ignition robotics the Lord and all these big projects in one slide. Hear what we try to do is to break up their robots after a problem into a bunch of small sub problems. So imagine like Luca was showing previously in that opens for skimmer projects. You have the device driver known which may be can get there all data from the images. Papa said in a standard form of a message that is a abstracted image and then

you can have further notes that listen to that message through some more processing pass on more messages and stuff. So when you build a robot Software System rather than building one gigantic, you know, robot binary, instead you end up lost and lots of lots of little small batteries that find each other. At one time. That's the Dynamic competition Graphics. She knows best represented by those colored bubbles in the middle. There in simulation is invaluable and Robotics because these systems cost a ton of money. So that the humanoid robot on the bottom there, you can see both of

the renderings and in the real world photo, that's the real nice robot. I don't even know how much it cost to sign up for many of them still sometimes and I can really put a price tag on if it falls over. It's very bad certificate. You know, what's on my calendar, local it hangs from the ceiling. So I can't really fall on his face, but yet you want to stimulate it, going up and down stairs, and doing all sorts of interesting, postures and nothing's. You are not willing to try on the robot until you're really. Is this going to work out? Cuz that's where it becomes extremely useful to

have a high-fidelity simulator. Also, on this particular example, that's a simulated Mars Landing site in a simulated, Mars Lander, obviously, those don't exist you. So when you're trying to send me a difficult, world are extremely expensive world so you can use this emotion from Works, to save a ton of time and money. In the hospital this is all irrelevant because Hospital facilities of course our critical National infrastructure everywhere they are and so you can't just run over to our experiments in them. If you want to see what happens with running different numbers of robots for

delivery or different sequencing of lifts or automatic doors, they know, how do you cure for a door? How do you queue for the list that you can't really run in an operational hospital? So that we can do as we can create models of these facilities in the algorithms intestine in this simulation? Sue to do that. We have been constructing. All sorts of stimulation assets. So they're all sorts of different roles of people in in Hospital. Of course, when you go, there's all sorts of different staff. Different visitors different patients with different Mobility. AIDS there to help them as

well as all sorts of hospital equipment. At it's amazing when you start walking around large hospitals, how many things there are? That are basic in the shape of a thing. On Wheels, you push around, its kind of astonishing everything from any kind of diagnostic equipment to base. Create lots of different shapes of computers on wheels that they pushed around cowls cows everywhere getting pushed around sanitation karts. That's just a few of them that we've modeled in the picture there that there's anything to start of the shape of a cart that you can push through a doorway, they've got

it. So we're modeling all these things so we can see many things and some of the Motions of the the algorithm sitter is going to drop the girl Bus Around Mechanical infrastructure by which we mean doors. And lips are a key part of this operation because large Urban hospitals are essentially stacked vertically, so it's very common for a large, Urban hospital to be 10 level on several levels, below ground, many levels above ground. Where most of the infrastructure to run the facility and to provide for the patients needs is in the absence of the basement levels. So things

like the Central Kitchen, in the central, Surgical Supply sterilization areas laundry in. Imagine how much longer is produced in a large Hospital, right? They're going all those women still get in the most and sanitized and ready to go back up. That's all I'm in the basement of the building and then all that stuff needs to be distributed up. Vertically, through all the patient words are used aboveground for no, airflow, Enlighten, all that stuff. So the vertical Transportation, provided by the elevators are really bad with dictate the throughput of this entire operation. So

giving the simulation of the elevator cabins or Lyft, cabins, corrects is is a key part here, modeling in the networking topology of the lifts in many of these hospitals, are quite sophisticated. No, use of Robotics. They have certain Banks of of lips are only for robots and then so you can schedule those totally for their for their growing operations, then his other Banks of elevators that are used for patients Other banks used for visitors other Banks. It's very complex in a typical of hospital will have thirty years or more list just in the, in the complex. As you can imagine, once the

facility is built, its approximately in front of expensive to add more lift. So this beginning the most throughput possible out of the current lift is, is required in order to keep introducing more gobus. So here's a few of the common health care of us today that you go, if you go to public hospitals here in Singapore, you'll see some or all of these. The the Panasonic house. Once they use swisslog meal cards simply that there will be no bitch at me and then the second one from the left here, it's a cart full of food that sits on top of a little robot that

goes underneath that card and lifted up a few centimetres, then I can drive it to where it needs to go. So I'll do, if you're, if you're awarded at a hospital here in Singapore of your lunch or breakfast or dinner, all comes on, these fish for Go Bus, the central kitchen is in the basement. They, they drive him around to the the service lift, and they bring the card up to the table where you are, they drop it near the, the nurse area in the ward. In the nurses, will then hand deliver the the trays of food out of them cart. You can imagine if that card is has the food for 40 people at very

heavy. You know, so it's turned down every breakfast, lunch and dinner, they have to distribute these cards to all the different words across the stage seven levels of patients. Then bring it back 2 hours later today, have all the dishes washed and ready to go. So now I'm going to talk a little bit about what do we do with this? Simulation model that we've created? And then the goal is to allow multiple robot fleets. So this simulation snapshot on, this is like, what would be great to happen in the future? Just as kind of a cartoon of you have three

different robots from three different vendors but they're sharing the same space and operating reason to be around people. So what we're trying to do is to have it so that you can mix and match these robots. You know, the question course is why we're doing this is just making everything more complex and that they kind of easy single side. Answer is that robot, it's a rapidly developing field know every year. If you go to robotics trade show, there's new applications. There's new providers of robots with new robots, of course, until today, we do a lot of delivery in the field. So you see

lots of like meal. Delivery Lennon delivery medication delivery case files in a hospital. All sorts of stuff moving up and down there on the facility. That's the deliveries are coming today but I think we can all agree that when you up locations in the future things that we can maybe see coming on the horizon or like more and more of automatic cleaning robots to imagine like the Roomba Vacuum but like you know, when we're supposed to get hit with more, maybe like a, the ultraviolet light so I can spray around to do the sanitation, sanitizer rooms, and I can scrub floors, my floors will not

be so cleaning, a certain income in security robots. You don't put rolling around when the facility supposed to be closed or not supposed to be someone there or if they're supposed to come in there or whatever. So all these applications are coming and then of course is many that we probably can't even imagine right now that will be out in a few years. So we want to have our software systems that allow new applications to come into the field and just apply them without having to come sit there and Riri plan and reschedule all the all the machinery. So, if you look at the picture is a

picture of committing another day, we took that picture. If you want to go just looking down from one of the big towers there, you can see there's an incredibly complex and Field Complex. My flow of traffic. There are as many different roadways. They're all kind of Come and Go in different directions in an outflow spine. And that's because there's a well, you know, well, understood the rules of the road that people follow most. And as long as most of vehicles, follow most of the rules most, the time traffic can flow. And when you introduce the den, plus one vehicle, it just merges

into this flow and eventually goes where it needs to go. And so our goal is to figure out something similar for robotics. So this is the situation today and this is been in one of the hospitals here in Singapore, the parts of the hospital that are really impressive. They they run essentially one robot Fleet and let the picture of like that on their upper right there. Then milk Arts. This is taking the take the food out of the essential Kitchen in the basement. Distributes it up to the wards and then like an hour later, it brings back the the dirty dishes and they said the giant facility that

wants his audition. It's impressive, these are large robots. Those are very heavy carts, full of food for 40, 50 people, you know, it's it's actually kind of amazing and almost hypnotized him to stand there and watch these giant things. Move around very well. Orchestrated, the rainforest have some the single vendor please. This is this is really impressive to see this today, what we are trying to push those this multi-vendor operations in the same space. So if you look at how things are kind of where it goes today, it's like the fleet's. Don't talk to

each other. So vendor a doesn't talk to vendor be and so then you have to either keep them safe. Printed in time like you don't you have the the meal robots, all go and then once more robots are back on the charger, then you can bring up the other like kind of miscellaneous deliver your mouth, that's one solution. You can also just hope for the best in going to just send them through. They all have course, have on board collision avoidance sensors, as many different modalities and Sylvan ever Collide. It's just the kind of do this kind of funny. Things to maswik Lodge. Robot dance. They

just look at each other in like one of them boxed up and coming, like it's for the Flyers to watch and eventually do all kind of the traffic. Also to allow that usually. So it's it's not often I think we can safely and so, wouldn't it be great if, if you can have a joint planner, that is planning for both fleets, or at least can kind of be like in traffic police. Like no, you stop. You go. That kind of thing. And finally did I mention a little bit before but the lift integration is key Associates hospital to have so much vertical motion happening and essentially it's a little bit hilarious,

I think about, but in some of these facilities, the reality is that There aren't enough service. Lift to keep dedicating new lifts, a new vendors mother was. So if you have two vendors of robots and you've allocated to Banks of services to those two vendors, actually can't bring on another robot bender cuz I don't have any more lists to dedicate to them since the infinitely expensive to add to his list in building. So what we're looking to do is to help multiplexes list so that you can have different vendors, share the same left, which is not current practice. How the golden is two more

specifically say in based multi Fleet conventions for for robot operation. So could we come up with something like a cross of Endor? Lane Network, for example, in these large building stuff, your vendor n + 1, you sort of download the roadmap. Just like if you have a self-driving car system, in your joining the public road, you download the public realm up and then, you know, she's going to drive anywhere. You can follow the existing traffic conventions. So this is a part of the project to be working on his two to develop some conventions like this that we couldn't push out of the

vendor-neutral soaps to our organization, doesn't make robots. We don't have anything about to sell so we're not turn them in a push one. Over the other what we're trying to do is instead of to generalize these ideas of train traffic motion. How do you call a Lyft? How do you open the door if I can have him? So, to be that we've been creating open source tools that you are out there on the upper left. This one is called traffic and odor. So the days we can get the floor plans from the facilities. So there's that's our, an office going to lay out a lower right there and you can see what

the annotated some robot Lanes on there. And then the upper guide is another ASAP another facility were with, with taking their blueprints to trace, the walls, the doors and stuff, and then added where we think they're about traffic, things should be. And you can see there's different colors in there. It's inside the different. Police might have different routes to the same space because there might be doing different Prospect. Bedside delivery as opposed to Surgical Supply delivery or what have you. An interesting thing, this is just a minute, resting Journey

wasn't going down. Is how much control do we really get over these robots? So these are all going to safety. Certified systems are operating in, in, in the hospital. So it's not like you can just go and casually run, some new software on these robots. They're basically, you know, completely welded shut. But instead what you can talk to you as a fleet manager API, and most of the vendors, provide, some of the symbols of an API in, where you can talk to some over, some, Perla, get your ass, maybe a sex, a Miller PC, what have you? Two to issue, various tasks. Or to get various information is

from So was one of them pool, maybe some cleats only will tell you where the robots are, but they won't ever. Let you issue commands to the robots or maybe other fleets will let you see what their go out to her and you can temporarily pause iRobot Roomba, just stop. And then you can say, resume your task and we'll keep going again, and maybe some other fleets will give you full control where you can like give it a path to follow up, you know, fall the path long as there's nothing in its way. And so the hope is that if we can

have a system which can deal with those various levels of control, then we'd be able to mix and match in a sense to do. What I would have traffic police officer does that when they stand in the middle of an intersection and just give hand signals to drivers to help the traffic clear out. So of course, the more control we have over these fleets, the more potential there is for improvement. So there's kind of a given take care of if if the realities is 3 / 146 their Fleet of robots. We don't get any control. We can just read where they are. Then okay, let's just reality. But we would

then suggested they have other fleets in the system that we could have more more control over. So, to do this work, reading this event or neutral system that the URL, it's that there and get him. And RMF portal has been away from work, this is the core scheduling system, so you can imagine that user interfaces, this case would be like that. The surgical supply staff making orders a request for the new supplies for that day's operation or the nursing staff wanting, an item delivered to the award that goes through a dispatcher at a splinter, which then talks to a negotiation process between

the different Fleet adapters and in a unified schedule, that can able to predict where the robots are going, how much time they're going to delay waiting, for this lift, it all, that kind of thing should have put that all into it, as Grand unified schedule of the facility of that. Some of the shared space will be allocated in a reasonable fashion. And one thing is kind of interesting about all these robots. Is they, all of them will yield to the people in the hospital. So, if you're, if you get in the way of a milk cart robot, it just stops. It's kind of funny then it'll like very

politely say like me know, please move aside, I'm trying to make a delivery so it'll then you move away and it just kind of keeps on this way. So essentially, the task that your prediction of the task completion time, it can it can always be delayed because maybe someone like left the box in the corridor, something in there. Go by was not willing to go around the block. So just stop there. It'll then ask the box to move aside over and over and over until they almost comical. He'll just like park in front of a box, asking the Box move side. Eventually it send the page during the operations

staff and then they'll come and move the box out of the way and they're going to keep going. So don't say anything, just ask him, be delayed arbitrarily and but they'll never go faster than you think they will. So it's a bit of an interesting planning problem. Robots, I'll have a speed limit on them around. Here is regulated select .7 liters per second or something must never go faster than that. It can only be delayed and see if you can imagine that does interesting things for the schedule. Right now you have to replant. The know, this girl bought is got there sooner than we

thought. So here's a little cartoon, so just do it was going to happen. So let's say that there are three different fleets of robots. There alternate use the same Corridor or the more likely in reality, the same lift Lobby to do, all the, all the past, and the kind of converge in the lobby, for Lyft with a going to bring to wait for Lyft car to come. So, it's interesting because the robots, even though they all are kind of the same, they're also two boxes on Wheels. They're all implemented differently inside. And so each plate, doesn't necessarily know what the other one is capable of one

might be a very tall, heavy robot where it has to move very slowly and turn very slowly. And other one might be more like a cylinder shape or content of turn on the dining lights faster and be able to maneuver more easily. So, each bleed smc's is just kind of generates what it would love to do in an empty world. If there were no other about fleas around, So that would be great but typically they're going to interfere in some sort. So then now whenever they can see what what the other fleets are trying to do then now you have a a game right now. What would I do if you do that, what you

do, if I do that and that game continues as deep as you have fleets and then at the end of it, you can now have a score and now a third-party judge can be the kind of arbiter between all those plans. And those plans are determined upon as an all-around account. And finally, we can pick the winner since of the winning plan. That requires the least for the joint delay on, everyone's desire stupid-face. That's what to do with fleets. It's also interesting to consider different classes of robots. So there are many robots that do not have a fleet manager shipped with them. So let's say,

you know, research projects, low-cost the graph, and prototyping projects, or even a standalone proof-of-concept or something like that. Where it's not necessarily a totally shrink-wrapped product that comes like with a fleet manager and technicians, to install to make it work. Great. And so, for those were having a fully open source fleet manager, which can plug in, at the same level of these proprietary Bender Fleet managers. So those hook into the common, internal robot software like Ross, for example, to use a mini research robots. And then it says, it's a peer of these

proprietary Fleet managers that she can have these mixed bleed. So you know, the police Minecraft robots as well as well as Randy or lower cross platform. So I will just first show what the simulation looks like. It'll come up a little bit. So here's for example, a one of the test facilities and one of the buildings that we working, this is designed to be kind of a mixture of a ward. So here's like killing people is a good for person over here as well as to stimulate some of the storage rooms and storage

facilities. Still like, this is what it looks like. When you go to the surgical supply store room through these little, like, lots and lots of shelving units with sterilize instruments that are in there in storage. As you say, patience. In transit, if you spend much time in the hospital, he's seen people moving around in various various ways. This over here is intended to the another kind of surgical store. oh, yeah, and of course, there's always tons of people waiting very first building

So what we can do is now also there that the traffic manager after I first introduced you to the robots. So this is a common robot that you can buy off the shelf nice robot made by a company called Mir. It's it's one of the common industrial sort of you want to move a thing. Type robot. This will move your thing that you can. You can put stuff on it or you can have accessories for that. Make the top lifts up and down to the other carts, the girls are nice and lower-cost platforms, and intended for rapid prototyping. And

that's he's so now we'll show him what happens when at the Grand planner all runs. So see her go somewhere else. Reduce type one handed. Look at this room is just going to go very fast. I'll pay the few times because one of the funny things with, with large robots because their speed limited to like point seven meters per second. If you watch them real time, it's like watching grass, grow right assistant. That's so boring or taking over the world has never seen real robots because they're so boring. They will

do it, extremely slowly, you know, it's fine, you're fine. They're friends of mine to spend the world of eight times here. So now this is showing that sing world but with him being around. So okay, missed up this way again cuz there's a lot going on on the right side. If you have that the top network from a girl wants, that was a screenshot from that traffic editor before. So here we have a purple Fleet there's a blue kind of Sion a fleet and orange Fleet the three kinds of robots here we're still only run into now but they're going to only move around and just like railroad tracks on

those on those pants. At any given time, the central planning entity is trying to predict the future of where these robots are going to go. So there's a green green arrow that you'll see in that green pad is the prediction. If you roll out the trajectory in the future like, where, where does it think? They're going out there going to be a stat time goes forward. Any given time? These little purple can of dots, and that's where the robots last reported position. So we're sending all this invisible boat with Road in reality as well in the same space. So the one the big white robot is going to

go from this simulated storage facility here to this simulated stores that sell to hear. He's just going to go back and forth these little robots here with your simulated, like a bedside delivery robot, you can deliver any item of the people or whatever. So those are going to be coming, just going everywhere. When they do is work with said the stuff on purpose, to be almost pathologically difficult because all of these guys are intersecting on purpose. So the central entities than going to be sort of stopping particular robots from from getting closer to Do you see them? Go back and forth?

They will stop and wait for each other. As the predicted path intersects over time you'll see some of them stop that goes to the be a situation here that we moved out of the way. It's it's sort of this like ballet as it goes on and you can imagine it's even more complex as we have more levels to it and more and more about the situation. So it's a little hard to see, what would have happened because we're not suing, what would have happened if you didn't have this sensual? Planner telling people to stop and start but believe me when I say that, it's not good, that it's just a train

wreck, just like, you know, just giant gridlock. That's what can happen with these robots in the goal is to make things available so that when you want to add and + 1 + + 2 robots, the future, it's no big deal. And so you can imagine a future world in a few years or who knows when when will be ten times as many robots as there are now where this type of assistant will become increasingly important to prevent these traffic jams from happening. Olympus pen play. It's going to be fun to watch.

so again it's almost like the software is running really well when there's no obvious traffic jams happening, which is a little bit you're going to like prove a negative but nothing you can imagine what would happen if if we just have the giant gridlock full of robots here, Listen to my prepared to talk here. These are all open source projects, there's some URLs and and get up there as well as the the kind of found a small project for Austin gazebo those are there with the websites that those happen we're always down for the Play. Welcome contributions from

everyone from all domains. It's a wide-open space the plane. So thanks for your time.

Cackle comments for the website

Buy this talk

Access to the talk “Open Source Robotics Middleware Framework for Healthcare”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free

Access to all the recordings of the event

Get access to all videos “FOSSASIA Summit 2020”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Ticket

Interested in topic “Artificial Intelligence and Machine Learning”?

You might be interested in videos from this event

November 6, 2020
Online
35
3.5 K
ai, amazon web services, apple, covid-19, deep learning, developing, devops, ml ops

Similar talks

Lakshmipathi Ganapathi
Software Engineer at Collabora
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Luca Della Vedova
Embedded Systems Engineer at Open Robotics
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Tern Poh Lim
Principal AI Consultant at AI Singapore
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free

Buy this video

Video
Access to the talk “Open Source Robotics Middleware Framework for Healthcare”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free

Conference Cast

With ConferenceCast.tv, you get access to our library of the world's best conference talks.

Conference Cast
892 conferences
35963 speakers
13720 hours of content