All right. Thank you guys for attending. I know I am right before your break, so I'll try to show cool videos and go over things quickly so that you guys can get your snacks. All right, so I'll start by saying that I'm an assistant professor in the School of Mechanical Engineering with a courtesy appointment in aerospace. And today I'll be talking to you about sensors, soft actuators and robotic systems. A lot of the stuff that the students in my lab we're working on now. So my lab is actually pretty interdisciplinary. We do sensor systems for a variety of applications. So we create sensor systems for robotics and system ID and other applications related to that, which I will be focusing for the most part on this talk. But I also had people in my lab. We work on optics, magnetics, radar, and of course applying these measurements to harsh environments. So the types of robotics that we do in my lab are related to soft and continue in robotics as well as actuators and sensors. We also do a bit of advanced interpretation. We also study a little bit about magnetic materials and magnetic sensing. There are a couple of my students who work on objects in radar techniques. One of my new students, I was actually working on a radar technique. And then of course we do a lot of interesting applications. We apply them to combustion environment, solid rocket propellants, hypersonic, multi-phase flows. So there's a lot of variety in my, in our labs work. I'll just focus today on the types of things that are probably most relevant to robotics that will be continue and soft robotic actuators and sensors. So I'll start with a little bit about modular continuum robotic actuator that we worked on in the past. The basic idea here is this is basically a snake robot. And the ideas that we have multiple modules that can actuate so that if you're doing a colonoscopy, for example, with an endoscope. You don't end up bending the colon, for example, you're actually conforming to the shape. So this is the skeleton, for example, of the endoscope that we've been developing. So it has a camera on the front. So we have specialized camera circuits. And each of the modules and maybe I can mouse. You can see my mouse, Yeah, Okay. So each of these are modules. In these modules have actuation systems in them, and they have two motors that apply tension to cables and they allow continuum bending motions. So I'll just start by showing some quick videos of how the robot works. So it's fairly modular. And each of these modules can bend in the x and the y direction. So you can see that for this video. And of course, because it is modular, were able to combine a bunch of them together, just plug them in together. And then the whole strand will turn on. And you can actually use that to do complex motions. So the motions that you would like to use for an endoscope, one of them maybe in an uncoiling motion so you can go forward. So I'll let it play out. And of course there are other motions that are important. Another one would be going around a corner, for example. So let's see. So in this case, this is a traditional endoscope with only one bending section and it will get stuck when it tries to go around corners. And it really does need external forces in order to help it bend. But using the multi actuator endoscope, we're actually able to create comfort conformations that are not possible to do if you don't have actuation. So I'll let this guy play out. Are. And what's interesting about this too is that it applies a lot smaller forces on all the walls when it goes around. For this particular project, we actually had to do some path planning as well. So we had to determine what pant plans we need to execute and what order and to make sure that there's some robustness in that. Of course, this is also just a generic robot. So we can do a lot of genetic robot things we can do scanning, we can look in different directions with the camera that's at the tip. And of course, we can do things that endoscopes aren't designed to do, like moving things around. And of course, if you use a couple of units, you can also do a little bit of crawling if you'd like. So it's, there's a lot of different things you can do once you have an interesting platform. For endoscope applications. We also wanted to do some sensing. So one of the sensors that we use, that's not sort of a traditional Center, are flexible Ben sensors. And that's actually something that a lot of my students in the lab work on together. They do a variety of different types of sensors. This one's a carbon black sensor. It changes its resistivity as a function of strain. And so we can map that out. And designer sensors, we put for them in a strand and put that inside the endoscope. We do a little bit of modeling control. And we do some modelling in order to get rid of some of the non-linearities. And then we're able to put them inside a sensor and then determine the configuration of the sensor. So here's one of the modules. We can put them inside the robot. And then what we're actually doing is measuring strain and using that to determine the bend angles on the robot. So we're actually doing, instead of doing the planning, we're doing the measurement and then we're doing the control based on the actuator. So we can do some small motions, we can do some large motions, and then we can do some simulations based on feedback control using sort of non-traditional sensors. All right, so once we have that put together, we can do just looking at the feedback sensors. On those sensors, we can actually use that to do a model or a simulation of what the conformation of the robot actually is. We can put all of that inside the robotic endoscope. And then of course, can put a sheath on it so that it can be used as an endoscope at the end of the day. Okay, so the next thing that I kinda want to cover for this particular project is that, you know, on top of that we can put together a user interface for it. And so that we can actually have people drive them. So this one has basically everything in it together. There's a simulation at the top. It's a little bit blurry. I guess this video's a little blurry. But the simulation at the top is using sensor signals and inputs to figure out what the conformation of the robot is in without having any sort of other visual feedback. So you can see that the simulation is matching the motion of the robot on the bottom. Okay. All right. So since this is sort of a, a, sort of an overview of the types of work that I do in my lab. I'll pivot a little bit over two from continuum robots, which are very difficult to model, to completely soft robots with materials that are compliant. So this is a completely different research area where we are studying the creation of compliant electromagnetic actuators. And these are developed using specialty materials. So we developed a liquid metal coils, we developed compliant magnetic composites, compliant iron composites, and some soft sensors in order to make a complete robot out of soft materials. So a lot of the times when you work with soft robotics, you find that the actuators themselves or traditional motors or pumps that are hard. And so if we want to make true actuators and two robots that are soft and we have to look at all the elements. So we formulated a bunch of different topologies. This is the first one that we're able to implement. And this one actually does inspired by the motion of Xenia coral, coral that underwater can pulse. And so we're trying to replicate some of the pulsing motion. So this is our first attempt at that worked out pretty well. It's basically a solenoid configuration. It has compliant magnets that are both squishy and magnetic. Would this play out? And of course, we have soft iron cores which are squishy. And they help you steer the magnetic field, but they themselves are not magnetic. And of course we have liquid metal coil, so these are gallium indium liquid metal coils inside silicon tubing. It's basically really high density liquid metal. So one of the things that we did was develop new methods for creating compliant materials. This is sort of one of the procedures that we use to create soft magnetic materials. And of course, once we assemble everything together, we have a solenoid that's capable of doing motions that are kind of small, but simulate the motions of a Xenia coral. What we can do sort of, once we have that kind of working is move on to a design that's a little bit more large stroke. So once we had a good idea of how things worked, we increased our stroke to 18 millimeters. And this is a very different design. It has flexures inside, so it's converting linear to rotary motion. And of course it's a completely different design on the inside as well. This is a solenoid. So this one can create much larger motions that are much closer to simulating the motion of a an actual Senior Corps. So I'll just let it play for a little bit. And of course is completely compliant. I might zoom to the end here. Where of course we're interested in sort of biomimicry. See maybe it's over here. I don't think this one has. Yeah. Okay. We're interested in biomimicry, of course, but we're also interested in doing motions like grit grasping, right? So soft grasping. You don't need to worry too much about contact points. It will just grab what you need. And of course, we can do all sorts of exotic shapes from very simple bokeh shapes to much more complex shapes as well. Alright. So sort of the last two things that I'll cover today are based on other types of soft robotic actuators. So this one is a MR. fluid based soft robotic actuator. Basically the idea here is we're going to use MR. fluids as a working fluid and we can turn on or turn off the flow or using different magnets. So we're actually using magnetic fields and soft actuators to control the flow of them are fluid and when you apply magnitude M R fluid, you'll freeze up and it'll create a valve. So in this particular case, my student Roman, was working on both hard versions of a bi-stable valve, so it only takes energy to turn on and off. And he also has soft versions of them are valve which are a little bit slower and has a lot of oscillations, but also work the same way. We can use this to control the flow of a fluid. But we can also use this to In a larger system. So at the very end of the day, we can apply it to new nets to use that as a control system for new net. So we have a soft actuator for our new net flow system. We can use that for grasping. And of course we can create our own grasshoppers that are magnetic. In this cases is MR fluid grasp or so has NMR fluid pumped into a cavity and it has a magnetic field on the opposite side to help you have a stronger grip. So there's a lot of different actuators we can apply in conjunction with the M R fluid based system. So in the interests of time, I'll go over the last soft robotic system that my students have been working on recently. So this is a pneumatic expand compress, a bend and twist actuator. The idea behind this actuators, It's got three pneumatic chambers. If you think about it, can bend, it can twist and it can orient. And I think the most interesting thing about this actuator is actually it's twisting mode. So these are the bending modes and actually can lift a lot of weight. And so I think the most interesting motion on this actuator is actually this twisting mode. And that is accomplished by having a constraint in the center so that the length can't change. But because you're trying to expand the actuators on the outside, it creates a twisting motion and you can use that to, for example, twist off bottle caps. So there's a lot of interesting emotions that you can accomplish with that. And of course, you have so many degrees of freedom that you can do grasping on top of moving, on top of crawling, all sorts of things with one module enrollments been working on sort of expanding this actuator to multiple modules so that we can do larger workspace types of motions. All right, so I guess what we'll do here is I'll quickly say that I didn't do all of this work myself. My students actually did most of the work. So I'd like to thank all the students in my lab, especially Roman and Noah for some of the work that they haven't in some of these slides. If you get a chance to talk to them about their research, that would be pretty interesting. I'm sure they'd be able to share much more stuff with you. Sebastian's my new PhD student in Robo, from robo. So he'll be working on some other cool things as well. So with that, I'll let you guys ask some questions and hopefully you'll get a chance to get a snack. So what Roman does on the direction of twisting as you change the order in which you inflate. So you inflate one and that'll bias that the rotation and then it will rotate in that direction. So if you turn on different one, it'll rotate the other way. The oscillation frequency or the speed or reduce molecule. So for us, this is a, so it depends on the actuator I showed you a bunch of different ones. For one of the ones that is maybe the most relevant is sort of the Senior Corps of the second Zeno coral. It's an electromagnetic actuator. So if we had a way to track the tip location, we can do for to control or reduce oscillations as needed.