**Advanced Puppeteering Techniques for Robot Performance**

The art of puppeteering has been elevated to new heights with the development of advanced robotics and artificial intelligence. One such innovation is the creation of robots that can mimic human-like movements, allowing for more realistic and engaging performances. In this article, we'll delve into the world of robot puppetry and explore the latest advancements in controlling robotic motion.

**Direct Puppeteer Control**

I have a whole different high-level set of controls that allow me to basically manipulate this and make it go from walking, which is where I'm 180 degrees out of phase, yeah, to starting to gallop, where the phase angle changes right. So, I have this one control that makes a big difference in how you perceive the motion. Yeah, yeah, those are the dials and the numbers, the sliders that you can adjust. But can you also integrate direct puppeteer? Yes, yes, I can do that. That's the cool thing; use the word layering a second ago, and that's actually a really good analogy. So, I have direct control over some of the parameters we're talking about, like this is that step length, and this is the height, so I can combine the algorithmic control of something really complicated with an expressive control that allows me to add what's important for the scene from an emotional standpoint without having to worry about all the technical stuff that's happening under the surface.

**Controlling the Robot**

And then I can have those things happening simultaneously. So, I'm controlling the physical location of the ball in space with an XYZ control, it's like a little joystick forward and back, left and right, up and down, and at the same time. The fingers are spinning, and the ball is moving; it's almost like the subconscious balancing that we would do in our normal gait or manipulation of objects combined with the very conscious then expressive manipulation of it. It's kind of magical honestly just watching it happen. Wow, that is so cool.

**Principles for Robot Puppeteering**

These principles are things that apply not only to access but also to other robots that you've worked on absolutely. I mean this type of thing applies directly to projects I'm working on right now, movies that are in production that I cannot even talk about but that allow me to have things like um like uh you know breathing and physical agitation and all the stuff that it's like oh let's bring the puppet to life. Well, I can bring the puppet to life in a way that's that's more compelling than I've ever been able to before now, and then do emotional puppeteering on top of that.

**Hades, the Robot Biped**

Let's talk about this thing for a second; yikes. So, Hades is a prototype robot from the Robot Combat League TV show, and it's also a walking robot. Axis over here is a pentapod; it has five legs, while Hades is a biped. The walking system for Hades is almost exactly the same as Axis, except the number of legs is two instead of five, and the system that controls the algorithm, the walking algorithm, is uh unlike Axis; you have direct control over the speed, you have direct control over the um direction right, the heading. I was talking about, and you also have direct control over the amplitude, which is how long the step is.

**Control System for Robot Combat League**

For something like Hades, the way the contestants would control the robots; they would steer it with basically a joystick, and they had this other control that would allow them to change their uh the amplitude of the step, and also make the robot crouch. Kind of like this control where I can bring it up and down, yeah; they could do that but with this robot. Imagine this robot going up and down, bobbing and weaving, the other thing that they could do was turn the speed down to zero and then just like crank on the amplitude, and the robot would snap into these fighting poses, it was really cool, it was also very unexpected.

**Collaboration with Concept Overdrive**

The control system for Robot Combat League was a collaboration with Concept Overdrive. The reason I did that was because they had everything to control hydraulic robots already figured out but the thing that they didn't have was the walking algorithm so we kind of had to figure that out together, um, but luckily I'd already made robots similar to Axis so we kind of knew what to do and whether it's a robot that's hydraulically driven or with electric servos; it's all the same principles in terms of the motion how you think about deconstructing the motions for animation for performance. Yes, it's decades of institutional knowledge and we'll include links and descriptions; it's for all places where people can find out more information about your projects.

**Conclusion**

Thank you so much for having us here, Mark. Thank you thanks for coming

"WEBVTTKind: captionsLanguage: enhey everyone norm from tested here and i'm so delighted to be in the workshop of mark zutrakian mark thank you so much for having us here thanks for coming it's really good to see you again norm it's so good to see you in this crazy year crazy time yes and you have been making robots for so long but we thought it'd be a great chance to come visit your shop for the first time to take a look at some of your old projects and just kind of dive into the world of animatronics and robot design absolutely well again thank you for coming uh and you know we've got i've got this robot here to show you this is axis and axis is a robotic art piece that i built in 2012 and the story with axis is that one of my colleagues chet czar who is an amazing artist himself he's an incredible creature effects artist but he's also a fine artist and he curates a show an art show a group show and he invited me to create a piece for the show and i had never really made something just for myself before in that way i've done you know basically commercial art for my whole career so i made a thing called cascade which is like a tentacle and there are a few recurring themes in my work one of them is tentacles and another one is mechanical hands so when it came time to do a sequel to cascade the idea that i thought of right away was something that was based on a hand a mechanical hand and that's what this is it's very specifically a sort of sinister looking hand with four fingers and a thumb and it's very gently and delicately spinning this globe and the thing that i really did not know how i was gonna pull off on this was the motion the the kinematics of how do you make something like this work with a ball and but i thought i could probably figure it out and i did and this is the result the movement is so fluid and the fact that you went for a hand which is something that you know we as humans we we wouldn't know this uncanny valley if it doesn't move naturally that seems like a an ambitious challenge but obviously you have years and years and decades of experience working creating robots and animatronics for the special effects industry this is directly interacting with a real object it's balancing this globe so can you break down what each finger is doing how it's constructed and what that animation is sure absolutely so the fingers well first of all the fingers are all the same they're all the same size they're all the same length which helps a little bit the other thing that i did was just to start thinking about this i came up with some limitations and the first was okay i'm going to define a plane and that's the base of the fingers i want to define another plane that's the plane that the fingers are touching the ball okay so i've got that and then i called upon high school trigonometry and i started figuring out the the angles these it's just a bunch of triangles so there's one triangle that describes the finger tilting on this this this axis like this there's another triangle that determines where this point is in space based on its height and then finally there's a triangle that determines what the angles of these two joints are to put that point where i need it and once i had that i was able to define a path in space and have these fingers walk along that path and when i use the word walk it's because it's basically a walking machine right it has a it has a walking gate it's a pentapod it's a five-legged spider lying on its back spinning a ball like a circus animal and that trigonometry what you solve for is a way to send instructions based on this design how many joints the distance between the joints so that you know where that tip of that finger is right at all times and then you can program an animation now is this we call it we call it inverse kinematics right so that's that's the familiar phrase in this case i'm using trigonometry to determine the inverse kinematics because luckily the fingers are reasonably simple i'm still learning i'm starting to learn a little bit about linear algebra and some of the much more complex ways determining inverse kinematics but this was kind of my baby steps into that field and i have to say that's one of the reasons why i do these projects is to really stretch and learn and grow as a as an artist and as a as an engineer is because i'm mostly self-taught is i'm constantly trying to come up with these challenges for myself to extend my work into other areas another thing i'll say is that i don't generally have time to figure out stuff like this on official projects so if a movie comes along the schedule is usually pretty short and there isn't a lot of time to experiment with inverse kinematics so i'd really do that on my own time and the benefit is that once i figure this out i can apply this to another project in the future um i don't know when i'm going to do a walking pentapod but you never know it might come in handy well in terms of the world of pentapods you didn't just build access it seems like you built this it does this very specific task but then you iterate it you made a next generation that's exactly right and what happened there was that i got a call from our friends at battlebots and they had seen axis and they said hey can we use it in the show we want to put the trophy on it and i told them that it was going to be too heavy so i made axis 2 and axis 2 has a couple of really key differences from axis 1. the main difference is that it can handle a lot of weight it can handle much more weight than than axis one could the servos are beefier the layout is more optimized for the physical strength the fingers are short and stout and the other thing that it has is it has optical stabilization there's a camera in the center of the palm pointing straight up that is right now looking at a black dot in the center of this paper and if i push this around the fingers will actually walk and pull that dot towards the center so this is how i'm able to put this really heavy trophy the battlebots trophy on this and have it spin for hours at a time and not fall off because it's actively centering the turntable and that's a much more ambitious challenge than holding something like a globe here it is and yeah exactly right because with axis one it's statically stable the ball just naturally sits on the points of the fingers it's like sitting on a tripod with this because this is a flat surface anything could happen this could this could slide over and it just did and now it's going to pull it back again and if you want to see how it works i've got a screen over here yeah that shows so it's computer vision it's it's computer vision yes you're taking an off-the-shelf kind of webcam exactly so this is a uh it's a webcam in the palm of the hand looking straight up and here you can see a little tiny dot and then here this is the area that i've zoomed in on and there's my dot and then down here the circle indicates that it's seeing the spot and it's focusing on it and it's actually determining two numbers one number is how far away is it from the center and the other is what's the angle so it's i call it heading and amplitude so the heading is this way and the amplitude is whatever it is and then based on that it will actually extend the legs towards that and pull it towards center because it's a it's a pre-calibrated system in the sense that the dot you know yeah is the center of gravity for this splatter this object that's even though the servos are the kind of advanced circles that have feedback yeah right but the feedback is what that's doing is it's um making sure that it goes where i'm telling it to go but i have to tell it where to go right and in order to do that i have to know this information the other thing that's interesting about this is that it works two ways meaning that i can go around the system and i can actually drive it so it is possible to take axis 2 and turn it over and drive it around like a walking machine because that same direction and heading it works both ways it works upside down and it works right side up right right and you've designed it so it is on a plane it is on a plane yeah although what i'd like to do in the future is get it to the point where it could basically handle any irregular object but for that i'm going to have to have force sensing exactly on the fingers and that's kind of the next the next big step for me i'm noticing on your interface here you have two waves here what do those represent so this is a a window into the gate generator and the gate generator is the thing that's basically making the thing take steps so let me see i've got a leg here somewhere oh there it is so with a leg like this there are two things that are happening one is that the i'll hold it like it's touching the globe one is that it steps off the globe and that's what's happening here every time you see this pop up the leg is coming away from the surface and then coming back down again and when it does that it's advancing advancing advancing and then it hits that moment and it retracts and it comes around so this is a step that has been deconstructed down to its simplest elements the foot comes off the ground the foot comes back down again the foot takes a step forward and then the foot takes a step back those two things in combination are the the primary building blocks of this simple gate generator then i multiply that by five and i change their phase relationship and it's basically a synthesizer and this is a recurring theme in my work now it's like a music synthesizer but it's synthesizes motion and each of in your quest you're layering these waveforms to create the the song of its movement essentially yeah which is the way you can think about you know when we walk as humans walking is a very complex thing for us it's simple because it's built into our lizard brain yeah yeah but you're deconstructing it into very discreet waveforms and then each waveform isn't just one axis of movement it is one range of motion yes yes that's right and it could be several well in the case of uh the step these all of these servos have to work in coordination for that to occur which is why we do the inverse kinematics all i care about is where the tip of this leg is what these motors have to do to put that leg some for some place in space i don't want to have to worry about that i just want to control the the business end of this leg and these motors have to follow along yeah so that's another thing that's really important about being able to write these control systems is you really take ownership of what's happening at a much lower level than than really i have in the past and when it's in this waveform format it's more manipulable than if it was a sequence of keyframes oh yeah absolutely the other thing is that since it's algorithmic i have um like you just said i have a different kind of control over it than i would if i was drawing it in an animation package because it's happening in real time right so my my puppeteering controls if you will are things like speed amplitude phase offset uh and these sound a little abstract but from a puppeteering standpoint if i'm controlling something um like let's say the scrunt and the m9 shyamalan's lady in the water the way i'm controlling it i'm not again i'm not worried about what the individual axes have to do i'm worried about the performance and so i have a whole different high level set of controls that allow me to basically manipulate this and make it go from walking which is where i'm 180 degrees out of phase yeah to starting to gallop where the phase angle changes right so i have this one control that makes a big difference in how you perceive the motion yeah yeah those are the dials and the numbers the sliders that you can adjust but can you also integrate direct puppeteer yes yes i can that's the cool thing you use the word layering a second ago and that's actually a really good analogy so i have direct control over some of the parameters we're talking about like this is that step length and this is the height and so i can combine the algorithmic control of like a running cycle something really complicated would be really really difficult to puppeteer and lay on top of that an expressive control that allows me to add the you know what's important for the scene from an emotional standpoint without having to worry about all the technical stuff that's happening under the surface and then i can have those things happening simultaneously so i can run the walking cycle oh my gosh and let's speed this up a little bit how about that wow so it's i'm controlling the physical location of the ball in space with an xyz control it's like a little joystick forward and back left and right up and down and at the same time the fingers are spinning the ball is this is almost like the subconscious balancing that we would do in our normal gait or manipulation of objects combined with the very conscious then expressive right manipulation of it right it's kind of magical honestly just watching it happen wow that is so cool and these principles are things that apply not only to access but for the other robots that you've absolutely i mean this this type of thing applies directly to i mean projects i'm working on right now movies that are in production that i can't even talk about but that allow me to have things like um like uh you know breathing and physical agitation and all the stuff that that it's like oh let's bring the puppet to life well i can bring the puppet to life in a way that's that's more compelling than i've ever been able to before now and then do emotional puppeteering on top of that yeah so that's it's really powerful or even for a layperson to manipulate if they're controlling you know a robot for a tv show that's yeah exactly so let's let's talk about this thing for a second yikes so this this is uh this is hades hades is a prototype robot from the robot combat league tv show and this is also a walking robot so axis over here is a pentapod it's got five legs hades is a biped the walking system for hades is almost exactly the same except the number of legs is two instead of five and the system that controls the algorithm the walking algorithm is uh unlike axis you have direct direct control over the speed you have direct control over the um direction right the heading i was talking about and you also have direct control over the amplitude which is how long the step is so for something like hades the way the contestants would control the robots they would steer it with basically a joystick and they had this other control that would allow them to change their uh the amplitude of the step and also make the robot crouch so kind of like this control where i can bring it up and down yeah they could do that but with this robot so imagine this robot going up and down bobbing and weaving the other thing that they could do and this was a real surprise was one of the contestants figured out that he could turn the speed down to zero and then just like crank on the amplitude and the robot would like snap into these these fighting poses it was really cool it was also very unexpected yeah and all this is the combination of the algorithmic gait that you develop combined with then letting people calculate different parameters yes of that in a way that's relatable and also right usable that's right super super i should mention that the the control system for robot combat league was a collaboration with concept overdrive the reason i did that was because they had everything to control hydraulic robots already figured out but the thing that they didn't have was the walking algorithm so we kind of had to figure that out together um but luckily i'd already made robots similar to axis so we kind of knew what to do and whether it's a robot that's hydraulically driven or with electric servos it's all the same principles in terms of the motion how you think about deconstructing the motions for animation for performance yes it's decades of institutional knowledge and we'll include links and descriptions it's for all places where people can find out more information about your projects thank you so much for having us here mark thank you thanks for cominghey everyone norm from tested here and i'm so delighted to be in the workshop of mark zutrakian mark thank you so much for having us here thanks for coming it's really good to see you again norm it's so good to see you in this crazy year crazy time yes and you have been making robots for so long but we thought it'd be a great chance to come visit your shop for the first time to take a look at some of your old projects and just kind of dive into the world of animatronics and robot design absolutely well again thank you for coming uh and you know we've got i've got this robot here to show you this is axis and axis is a robotic art piece that i built in 2012 and the story with axis is that one of my colleagues chet czar who is an amazing artist himself he's an incredible creature effects artist but he's also a fine artist and he curates a show an art show a group show and he invited me to create a piece for the show and i had never really made something just for myself before in that way i've done you know basically commercial art for my whole career so i made a thing called cascade which is like a tentacle and there are a few recurring themes in my work one of them is tentacles and another one is mechanical hands so when it came time to do a sequel to cascade the idea that i thought of right away was something that was based on a hand a mechanical hand and that's what this is it's very specifically a sort of sinister looking hand with four fingers and a thumb and it's very gently and delicately spinning this globe and the thing that i really did not know how i was gonna pull off on this was the motion the the kinematics of how do you make something like this work with a ball and but i thought i could probably figure it out and i did and this is the result the movement is so fluid and the fact that you went for a hand which is something that you know we as humans we we wouldn't know this uncanny valley if it doesn't move naturally that seems like a an ambitious challenge but obviously you have years and years and decades of experience working creating robots and animatronics for the special effects industry this is directly interacting with a real object it's balancing this globe so can you break down what each finger is doing how it's constructed and what that animation is sure absolutely so the fingers well first of all the fingers are all the same they're all the same size they're all the same length which helps a little bit the other thing that i did was just to start thinking about this i came up with some limitations and the first was okay i'm going to define a plane and that's the base of the fingers i want to define another plane that's the plane that the fingers are touching the ball okay so i've got that and then i called upon high school trigonometry and i started figuring out the the angles these it's just a bunch of triangles so there's one triangle that describes the finger tilting on this this this axis like this there's another triangle that determines where this point is in space based on its height and then finally there's a triangle that determines what the angles of these two joints are to put that point where i need it and once i had that i was able to define a path in space and have these fingers walk along that path and when i use the word walk it's because it's basically a walking machine right it has a it has a walking gate it's a pentapod it's a five-legged spider lying on its back spinning a ball like a circus animal and that trigonometry what you solve for is a way to send instructions based on this design how many joints the distance between the joints so that you know where that tip of that finger is right at all times and then you can program an animation now is this we call it we call it inverse kinematics right so that's that's the familiar phrase in this case i'm using trigonometry to determine the inverse kinematics because luckily the fingers are reasonably simple i'm still learning i'm starting to learn a little bit about linear algebra and some of the much more complex ways determining inverse kinematics but this was kind of my baby steps into that field and i have to say that's one of the reasons why i do these projects is to really stretch and learn and grow as a as an artist and as a as an engineer is because i'm mostly self-taught is i'm constantly trying to come up with these challenges for myself to extend my work into other areas another thing i'll say is that i don't generally have time to figure out stuff like this on official projects so if a movie comes along the schedule is usually pretty short and there isn't a lot of time to experiment with inverse kinematics so i'd really do that on my own time and the benefit is that once i figure this out i can apply this to another project in the future um i don't know when i'm going to do a walking pentapod but you never know it might come in handy well in terms of the world of pentapods you didn't just build access it seems like you built this it does this very specific task but then you iterate it you made a next generation that's exactly right and what happened there was that i got a call from our friends at battlebots and they had seen axis and they said hey can we use it in the show we want to put the trophy on it and i told them that it was going to be too heavy so i made axis 2 and axis 2 has a couple of really key differences from axis 1. the main difference is that it can handle a lot of weight it can handle much more weight than than axis one could the servos are beefier the layout is more optimized for the physical strength the fingers are short and stout and the other thing that it has is it has optical stabilization there's a camera in the center of the palm pointing straight up that is right now looking at a black dot in the center of this paper and if i push this around the fingers will actually walk and pull that dot towards the center so this is how i'm able to put this really heavy trophy the battlebots trophy on this and have it spin for hours at a time and not fall off because it's actively centering the turntable and that's a much more ambitious challenge than holding something like a globe here it is and yeah exactly right because with axis one it's statically stable the ball just naturally sits on the points of the fingers it's like sitting on a tripod with this because this is a flat surface anything could happen this could this could slide over and it just did and now it's going to pull it back again and if you want to see how it works i've got a screen over here yeah that shows so it's computer vision it's it's computer vision yes you're taking an off-the-shelf kind of webcam exactly so this is a uh it's a webcam in the palm of the hand looking straight up and here you can see a little tiny dot and then here this is the area that i've zoomed in on and there's my dot and then down here the circle indicates that it's seeing the spot and it's focusing on it and it's actually determining two numbers one number is how far away is it from the center and the other is what's the angle so it's i call it heading and amplitude so the heading is this way and the amplitude is whatever it is and then based on that it will actually extend the legs towards that and pull it towards center because it's a it's a pre-calibrated system in the sense that the dot you know yeah is the center of gravity for this splatter this object that's even though the servos are the kind of advanced circles that have feedback yeah right but the feedback is what that's doing is it's um making sure that it goes where i'm telling it to go but i have to tell it where to go right and in order to do that i have to know this information the other thing that's interesting about this is that it works two ways meaning that i can go around the system and i can actually drive it so it is possible to take axis 2 and turn it over and drive it around like a walking machine because that same direction and heading it works both ways it works upside down and it works right side up right right and you've designed it so it is on a plane it is on a plane yeah although what i'd like to do in the future is get it to the point where it could basically handle any irregular object but for that i'm going to have to have force sensing exactly on the fingers and that's kind of the next the next big step for me i'm noticing on your interface here you have two waves here what do those represent so this is a a window into the gate generator and the gate generator is the thing that's basically making the thing take steps so let me see i've got a leg here somewhere oh there it is so with a leg like this there are two things that are happening one is that the i'll hold it like it's touching the globe one is that it steps off the globe and that's what's happening here every time you see this pop up the leg is coming away from the surface and then coming back down again and when it does that it's advancing advancing advancing and then it hits that moment and it retracts and it comes around so this is a step that has been deconstructed down to its simplest elements the foot comes off the ground the foot comes back down again the foot takes a step forward and then the foot takes a step back those two things in combination are the the primary building blocks of this simple gate generator then i multiply that by five and i change their phase relationship and it's basically a synthesizer and this is a recurring theme in my work now it's like a music synthesizer but it's synthesizes motion and each of in your quest you're layering these waveforms to create the the song of its movement essentially yeah which is the way you can think about you know when we walk as humans walking is a very complex thing for us it's simple because it's built into our lizard brain yeah yeah but you're deconstructing it into very discreet waveforms and then each waveform isn't just one axis of movement it is one range of motion yes yes that's right and it could be several well in the case of uh the step these all of these servos have to work in coordination for that to occur which is why we do the inverse kinematics all i care about is where the tip of this leg is what these motors have to do to put that leg some for some place in space i don't want to have to worry about that i just want to control the the business end of this leg and these motors have to follow along yeah so that's another thing that's really important about being able to write these control systems is you really take ownership of what's happening at a much lower level than than really i have in the past and when it's in this waveform format it's more manipulable than if it was a sequence of keyframes oh yeah absolutely the other thing is that since it's algorithmic i have um like you just said i have a different kind of control over it than i would if i was drawing it in an animation package because it's happening in real time right so my my puppeteering controls if you will are things like speed amplitude phase offset uh and these sound a little abstract but from a puppeteering standpoint if i'm controlling something um like let's say the scrunt and the m9 shyamalan's lady in the water the way i'm controlling it i'm not again i'm not worried about what the individual axes have to do i'm worried about the performance and so i have a whole different high level set of controls that allow me to basically manipulate this and make it go from walking which is where i'm 180 degrees out of phase yeah to starting to gallop where the phase angle changes right so i have this one control that makes a big difference in how you perceive the motion yeah yeah those are the dials and the numbers the sliders that you can adjust but can you also integrate direct puppeteer yes yes i can that's the cool thing you use the word layering a second ago and that's actually a really good analogy so i have direct control over some of the parameters we're talking about like this is that step length and this is the height and so i can combine the algorithmic control of like a running cycle something really complicated would be really really difficult to puppeteer and lay on top of that an expressive control that allows me to add the you know what's important for the scene from an emotional standpoint without having to worry about all the technical stuff that's happening under the surface and then i can have those things happening simultaneously so i can run the walking cycle oh my gosh and let's speed this up a little bit how about that wow so it's i'm controlling the physical location of the ball in space with an xyz control it's like a little joystick forward and back left and right up and down and at the same time the fingers are spinning the ball is this is almost like the subconscious balancing that we would do in our normal gait or manipulation of objects combined with the very conscious then expressive right manipulation of it right it's kind of magical honestly just watching it happen wow that is so cool and these principles are things that apply not only to access but for the other robots that you've absolutely i mean this this type of thing applies directly to i mean projects i'm working on right now movies that are in production that i can't even talk about but that allow me to have things like um like uh you know breathing and physical agitation and all the stuff that that it's like oh let's bring the puppet to life well i can bring the puppet to life in a way that's that's more compelling than i've ever been able to before now and then do emotional puppeteering on top of that yeah so that's it's really powerful or even for a layperson to manipulate if they're controlling you know a robot for a tv show that's yeah exactly so let's let's talk about this thing for a second yikes so this this is uh this is hades hades is a prototype robot from the robot combat league tv show and this is also a walking robot so axis over here is a pentapod it's got five legs hades is a biped the walking system for hades is almost exactly the same except the number of legs is two instead of five and the system that controls the algorithm the walking algorithm is uh unlike axis you have direct direct control over the speed you have direct control over the um direction right the heading i was talking about and you also have direct control over the amplitude which is how long the step is so for something like hades the way the contestants would control the robots they would steer it with basically a joystick and they had this other control that would allow them to change their uh the amplitude of the step and also make the robot crouch so kind of like this control where i can bring it up and down yeah they could do that but with this robot so imagine this robot going up and down bobbing and weaving the other thing that they could do and this was a real surprise was one of the contestants figured out that he could turn the speed down to zero and then just like crank on the amplitude and the robot would like snap into these these fighting poses it was really cool it was also very unexpected yeah and all this is the combination of the algorithmic gait that you develop combined with then letting people calculate different parameters yes of that in a way that's relatable and also right usable that's right super super i should mention that the the control system for robot combat league was a collaboration with concept overdrive the reason i did that was because they had everything to control hydraulic robots already figured out but the thing that they didn't have was the walking algorithm so we kind of had to figure that out together um but luckily i'd already made robots similar to axis so we kind of knew what to do and whether it's a robot that's hydraulically driven or with electric servos it's all the same principles in terms of the motion how you think about deconstructing the motions for animation for performance yes it's decades of institutional knowledge and we'll include links and descriptions it's for all places where people can find out more information about your projects thank you so much for having us here mark thank you thanks for coming\n"