Robotics Engineer Builds a Proper Robot Arm in Scrap Mechanic!

[@blueflame_sm]

My guess is that real life robots abstract away the range limitation by using only positive coordinates. Internally it probably still uses itself as (0,0). In trigonometry there is this fun arctan2 function, which does the same as tan, except it takes 2 inputs (y, x) and outputs the angle corrected for quadrant.

[@kANGaming]

hmmm atan2 would be the easiest way to do it. I wonder if the modpack has it or not

[@blueflame_sm]

@@kANGaming scrap mechanic's lua api supports it, and i'm pretty sure the modpack implemented all of scrap mechanics math functions. (it may be built into the atan function if its supplied with 2 inputs, cuz atan2 with inputs x, 1 is basically just atan)

[@robertfield7532]

@@kANGaming no idea about scrap mechanic, but can confirm atan2 is the way to go

[@vibaj16]

Well atan2 just does what kAN did. I don't think there's really a better way to do it than the conditional

[@Rocketenginetoaster]

​@@kANGamingarctan2 is in the math block

[@robertfield7532]

As a recently graduated engineer, this tickles all of my feathers

[@anteshell]

As I failed my mechanical engineering degree due to health issues I never received my feathers, but my interest towards these never died so this surely scratches my balls right from where it feels good.

[@Gentalkat11]

Yo what

[@azman1271]

You're a furry?

[@Ekipsogel]

As someone who hasn’t finished 8th grade, I’m lost.

[@darealkosmo]

@@anteshell💪💪

[@stanislaw.wokulski_]

Hey Kan, your work on scrap mechanic and other mechanical games inspired me to learn about roboticcs, and it turned out to be my passion. i just got accepted to university to study robotics, so i just wanted to say thank you.

[@thomasmclean9406]

Nice work man!

[@Rocketenginetoaster]

Good on you, I'm going into my final year of secondary school soon, if all goes well I should be able to get into engineering A Levels

[@Poldovico]

"We're gonna do all the math, I'm so excited" That's how you know you're listening to an engineer.

[@vibaj16]

Or a mathematician

[@AlechiaTheWitch]

​@@vibaj16they usually dislike math Why do you think they find all the shortcuts to do it

[@dopebossgamerz]

​@ttry1152 you like playing videogames but are excited when you find a cool shortcut

[@vibaj16]

@@AlechiaTheWitch Engineers usually dislike engineering. Why do you think they find all the shortcuts to do it?

[@kANGaming]

I like good maths

[@planetearth8044]

You can really tell how much he genuinely loves everything in this video, robotics and math and such It improves the experience so much hearing him exited about it

[@Thagrynor]

God, I really hope you do a ton more of these videos. Love learning more about that field and hearing your experience with it. Seeing it done in SM is just the icing on the cake. Keep it up dude. You could make it large enough to put a spud gun on the end and use it to automate carving something out of Cardboard blocks.

[@nutmeggaming11261]

One of the things I love the most in life is hearing people love something and hearing them talk about. Please keep making these, I want to know everything about what you did and what you enjoyed.

[@mrbeeznz5196]

As a mechatronic engineering student I shall be watching this video with harsh eyes

[@jakobnorrestam]

This is that thing I've never seen other scrap mechanic youtubers do, and I absolutely love it. I love every video where you apply actual real life engineering to your creations, whether it's doing a bunch of math with logic gates to control that massive elevator you did once, or making an actual mechanical transmission even though it obviously is quite bad due to scrap physics, the 4wd piston car which was even more terrible due to lag, or any of the 100 other creations you have done like this. You get the point, I absolutely love everything that has to do with mechanical engineering, programming etc. Thats what sets you apart from all the other scrap mechanic youtubers, what makes you unique, and thats why I like watching you the most. I think you're the best, even though the numbers maybe havent shown it throughout the years, but thats because that target audience is unfortunately smaller. I would absolutely love further videos of this, and I also miss the streams you used to do!!

[@vit.budina]

As you said, the Z for altitude is the standard in most fields, most notably CNC mills and 3D printers. Lathes use X as cross feed and Y as longitudinal feed, and in computer graphics, Y is used in place of Z for altitude, presumably to make it easier to transition from 2D to 3D, keeping the altitudinal axis the same in both dimensional workspaces.

[@tacoterrorizer1862]

The reason robotic arms use z as the up axis is because they align coordinate systems at each joint, with the z axis being aligned along the axis of joint rotation. Since the first joint usually rotates the entire robot, z is always aligned upwards

[@Trave1s_with_Chris]

Also, in real world applications z axis should always be used to represent vertical axis as x & y are 2 dimensional. And in the context of 3 dimensions z is vertical, t is the 4th but it gets much more complicated once that is added.

[@invention64]

It's generally just a mistake made by computer programmers when implementing 3D. I'm guessing it has to do with y being the vertical axis in 2D computer graphics, which then you might make the altitude in your 3D engine. It also makes it easier adapting 2d jumping code to a 3d game if you keep y as altitude.

[@endwiserd1234]

I missed when you did cool stuff like this and streams when you were working on this kind of stuff

[@ericsternwood9812]

I'd love to hear those stories! along with anything related to math and physics.

[@nessa6859]

This is actually really interesting for me, I used to work on the line at GM, and there were robots similar to this where I clicked a button and the robot moved from position to position placing a part so we could then drill it in place. Would love more robot stories and videos.

[@kANGaming]

I also worked at GM for a while :P

[@nessa6859]

@@kANGaming ahh we we're co workers at some time then maybe :P I live in the same area as you

[@JanTonovski]

This is really cool kAN! I would love to hear more about the robotic stories! I am looking to study aerospace engineering after I graduate from highschool

[@bengineer8]

Very cool and easy to understand. Yes, more robot stories!

[@calvin_bramble]

Love the passion that is coming out in this video. I am currently working towards my own engineering degree and was geeking out *hard* the entire time, looking forward to the next robot arm :)

[@MaddaloniBruno]

I don’t comment often on stuff on the internet, but I have to say - I love your enthusiasm about this topic and it’s so much fun watching you create these robots and explain it!

[@karaolli6430]

I think reenactment of the stories that you want to tell us in Scrap Mechanic might be an interesting idea.

[@moddien1]

Would love to hear these robot stories :) Have you played Space Engineers? It has block welders and merge blocks. Pretty sure you could build a whole assembly line in there.

[@kANGaming]

I have but never really made too much automation in it. The logic in Space Engineers is just programming in C which is less exciting than trying to do stuff with hardware

[@moddien1]

​There are timer blocks for simple things that can be used as logic gates, and they recently added the event controller which lets you automate stuff via gui. For the multi-axis robot you had to use the C programming though and Scrap Mechanic is just superior for tinkering. Keep on with the good content, I really enjoy it.

[@MrMattkid]

For your variable speed problem, all you need to do is get the ratio in distance for each axi. So first you need the distances | X1 - X2 | , | Y1 - Y2 | , | Z1 - Z2 | . Next, you get the ratio for them, I would either pick the longest or shortest distance as your base, then the other you divide by your base. The base would move at your predetermined speed. If you want that speed to be your max speed, then use your longest distance as your base, short_dist / long_dist, and middle_dist / long_dist That'll force the other 2 to always be a fraction. For their speed it's as simple as multiplying those fractions with your base speed. Slower = middle_ratio × base_speed and slowest = short_ratio × base_speed. I hope that this reaches you before you try and figure everything out. Good luck.

[@kANGaming]

This is kind of close but only works if the robot has symmetrical arms. The way I would do it is look at current angle for each axis, next angle and then divide all by the same speed so they move across each angle. Hard to explain but the issue is each rotation needs to move from A to B in the same amount of time I think in order to create a linear movement.

[@MrMattkid]

@kANGaming I guess do something similar but using the difference in angles for the two positions instead of the difference in coordinates.

[@MrMattkid]

​@kANGaming So using your naming convention for the variables, the part at the beginning of the calculation instead of the x, y and z positions you'll use the angles instead, so it'll look like this: |

[@darealkosmo]

Nice one bro! Looking forward to seeing where you take this robot arm series :)

[@trerobinson4733]

This is the kind of stuff I want to do for a career work on the automotive and factory robots

[@creeperx3sssboom347]

I wish you luck in your endeavors!

[@Jarko1401]

Really cool video. As a programmer i sometimes like to create simple CPUs in games, so I understand what you feel buliding this. Looking forward to more DoF.

[@jdwg5]

Hey @kANGaming! Fellow robotics engineer here! I did a capstone project at University on developing the IK for a 6DoF robot arm, so this is a fun throwback! Quick note on the -X issue for your joint one solution: the way I handled that edge case was by putting in a (X/|X|) term in front of the sin(Y/H) term to get your J1 angle, that way there's no logic and it's purely math based! Hope this helps!! Also, the real pros don't have to worry about that edge case cause the proper way of deriving the IK solutions is to solve using the coordinate transforms and such, those will end up giving the proper solution to that problem with much fewer edge cases.

[@dinoduck5121]

Would love to see more stuff like this! It’s so interesting to learn about, and it’s really cool to see you so excited about this stuff

[@neko7684]

The only robotics experience I've had is a basic remote control robot, and software wise my two 3d printers. Somehow I still managed to understand how this works. Good job on explaining it!

[@kjdh3823]

This is amazing man! I was so bad in maths at school but I totally got this stuff now…💀 Would love to see more of this! This is the way I would say that also adults can/should play games. This is educating, fascinating and fun.

[@aprcktiplaal9293]

kan explaining some simple trigonometry as if its rocket science lol but i get that hes just erally interested in it and super invested, its nice to see people so happy about their hobbies/past ^^ also for the moving at the same speed problem just take each motors current angle, and each motors target angle get the difference, and that should technically be yourspeed lets say this: Xc (Current angle), Xt (Target angle), Xd (Angle difference) Ac = 1, At = 5, |Ac-At| = Ad = 4 Bc = 14, Bt = 27, |Bc-Bt| = Bd = 13 Cc = 28, Ct = 54, |Cc-Ct| = Cd = 26 Dc = 14, Dt = 27, |Dc-Dt| = Dd = 13 this works, assuming that speed 4 is twice as fast as speed 2, aka its linear. this works because if speed and distance are the same, then all distances should be traveled by a unit of 1 for example a car driving 100 km/h travels 100 km in one hour so lets say a second car has a route that goes 60 km if it trives 60 km/h it will also arrive at the destination in 1 hour. also im wondering aobut the 6/7 axis robot arm will the wrist, when it rotates, make all the other arms adjust for the length of the wrist/the position on the tip? or will the wrist be a stationary point determined by XYZ, and the wrist just goes from there and extends into whatever direction you give it? also you need to consider the fact that you can enter coordinates/angles that will make the robot intersect with itself. good luck with the logic, and keep up the good vibes!

[@tacoterrorizer1862]

On a 6+ axis robot the entire robot adjusts its joints to keep a pint at the end of the wrist in a single location. Usually there's a tool on the end of the arm so it can be adjusted more to keep the tools center point stationary while the entire robot rotates

[@aprcktiplaal9293]

@@tacoterrorizer1862 ah, makes sense, so theres a bit more logic to be considered when designing a 6+ axis robot arm then.

[@tacoterrorizer1862]

@@aprcktiplaal9293 yep, the extra 3 axis allow for the end point to be rotated while holding constant position, meaning you have to define x, y, z, rx, ry, and rz for all taught points. A 7th axis allows for dead space and gives the robot infinite solutions to reaching a given point, making it even more complex

[@tschuschka]

I absolutely love the fact that I know all the math from high school, makes me feel good about the college course :)

[@como42]

I'm 30 about to turn 31 and looking into engineering as a goal for my career. This piques my interest just a little bit more in that direction. I appreciate you giving your backstory along with the explanations. I look forward to returning to this when I get my feet under me a little more. I'm always gonna keep trying.

[@aprcktiplaal9293]

also, you could use a vacuum pump part on the end and use it to place blocks and perhaps build something like a complicated 3d printer

[@bradwb6192]

So glad to see you using all that knowledge for something you can be excited about. Thanks for sharing...

[@nathaneveleigh7848]

As someone who aspires to go into Robotics, I'd love to see more of these!

[@Nevir202]

This is cool. Even if you can't find a paint or some kind of construction mod, I'd like to see you build a robot that can autonomously stack blocks. Just that one project could get as complicated as you want, for example, if you wanted to make it perform a grid search (should be possible with a set of memory modules, latches and a distance sensor) to figure out where the blocks are to start with, then stack them in an arbitrary location that you input.

[@hectormidgley5641]

I just love watching you explain everything about robotics. Can't wait for you to put a flamethrower on the end of a scrapmechanic robot arm and tell the story of the flammable paint.

[@azlan-e9s]

I love seeing you so existed it makes me so happy

[@isaacspencer96]

This is AMAZING!! I love robotics and am about to go into college for robotics engineering technology. That math breakdown section was really helpful, I was literally just trying to make one of these in scrap mechanic the other day!! I didn’t get anywhere near this far though… Anyway, great vid, PLEASE upload more of this stuff!!

[@djah0063]

Did... did you really do inverse kinematics in Scrap Mechanic? As a current automation engineer, bravo.

[@Blueberry_12000]

i just recently learned sine, cosine, and tangent so this doesn’t make much sense but u still enjoyed you expressing you love for robotics

[@zw345]

I have never thought about all the math behind these things. I worked with the 6 and 7 axis arm style and 5 axis gantry robots in the molding division of a automotive supplier. Most of the time it was fine tuning where the robot stopped to either interface with the parts, another tool, or placing the parts on a belt. I've had my fair share of damaged EAOT's to repair after an unscheduled stop into a nearby nonmoving object.

[@aabuyaa]

some of your videos are literally podcasts and i kinda liked that. as an ex engineering student it was fun to listen in.

[@mrbeeznz5196]

Also I want more of this, definitely down my alley, might even make my own

[@triviavanille62]

this robot brings back memories of my last job... we made fiberglass doors, and the process to make them is to use a hydraulic press to mold and cure the raw fiberglass to form skins for the main panels of the door, then attach them to wooden frames and add pressboard fillers on the edges. when the skins came out of the press, they were picked up by this massive 8-axis robot that then carefully held the skin over a spinning blade to precisely trim off the excess material. it was insane how precise those robots were, and somebody had to program all the different door models into the software so we just pushed a couple buttons to select the proper model number for the robot to trim it properly

[@epicwolfboy11d3]

I am a senior in High school and I want to become a robotics engineer, I would love to hear some of your stories to see what I have to look forward to in the future.

[@Elucidus4]

Glad you are so passionate about it, also glad I could follow along at 1.5 times speed. Trying to get my kids interested in robotics.

[@Arthur-xu2iy]

I am interested in robotics for some time now, i would love to hear your storys Kan! These robots look amazing

[@Hadeks_Marow]

Ok, so I am a programmer. 3D space for animation/simulation in this case. What I find interesting is when you are making a robot arm, you are solving for angles of B, C, D where the positions sort themselves out based on the rotations. When I do it however, (due to the differing nature of what I do), I have to solve for the 3-dimensional positions of points B, C and D, where from there, it auto-gets the angles by telling point-B to look at position-C and point-C look at position-D by converting the vertex to localize the grid direction. I cannot tell which method is harder. But strictly working fpr the case of 3d virtual simulation standpoint, working with rotations is a very hacky solution rather than defined points as that tends to be more absolute when it comes to 3D simulation on a computer.

[@vibaj16]

I would say your method is worse in this case, but not necessarily "harder". I think kAN's method would end up being more precise than yours when simulated (I think both are exact in theory, but the finite precision of the computer's calculations would be a slightly bigger problem with your method). However, unlike kAN's method, your method can be extended to robot arms with more than 1 "mid-joint", but in that case you have to iterate multiple times and even with infinite precision it won't be exact with any finite number of iterations (in most cases).

[@Hadeks_Marow]

@@vibaj16 I. . . don't think you get how vector3 vertex calculations work nor the context of when you would actually use them. :/

[@vibaj16]

​@@Hadeks_Marow I think I do, but we're probably misunderstanding each other if you get that impression from my reply.

[@Hadeks_Marow]

@@vibaj16 and thus I still have to disagree with that thought. Those that "may" have a false understanding will always think they have a proper understanding. Self-validation is by no means a form of actual validation I'm afraid.

[@vibaj16]

@@Hadeks_Marow I really do have the understanding tho. Maybe you can actually explain your thoughts instead of trying to guess how well I understand the subject matter.

[@elTT91]

Hey Kan, good to see your "engineering style" videos again ! It's really well explained with "simple" maths. And on the top of that, your video could inspire young kids to find their dream job !

[@MadMikael]

This is so great, I followed most of it. Can't wait to see on how you improve it... and even moreso how you can find a use for it in SM. Also hoping for robotics stories.

[@germandkdev]

That was a great explanation of the reverse kinematics Kan. I would love to see an actuator at the end, maybe just a simple grabber. Also different movement besides linear movement could be interesting too

[@TheObliviousF]

I dont know trig math too well, but now i want to relearn it just because of how cool robotics are, also would gladly listen to any stories you have from your time working with these robots.

[@tommykj2]

I would love to see you take this concept and build a tank with it. I'm sure there's lots more to it, but adding a ballistic solution to finding the end plot doesn't seem like it would be too far from where you're at with this. I'm not sure if the game offers the right sensors, but to add in main gun stabilization while moving and staying on target would be even cooler. I was an Abrams tank systems maintainer, and it's ability to stay on target while moving over most any terrain always fascinated me.

[@misapoisonify]

Love this video so much, currently going to college for mechatronic engineering and want to work on robots myself and seeing the math behind the robot just shows me an actual application to the math I am learning

[@goop9086]

Its realy cool seeing you make these more complicated machines. Its also realy cool when you get exited about your machines

[@BazilRat]

"Robots with kAN" sounds like a neat new series I'd watch!

[@IgnavumFortuna]

those are some flashbacks to my robotics lectures. exiting stuff.

[@EmeraldPencil46]

This is amazing, especially in Scrap Mechanic lol. I can actually understand what's going on, and it makes sense. I'd love to see more of the progress of this robot.

[@alexbrewer9930]

When we had an arm setup like this, it only took polar coordinates + z-axis (r, Theta, z). It threw plenty of people off because we’d all done work on the CNC mill before and were used to the Cartesian (x, y, z). Half the people (myself included) had already written code that had to have all the points converted.

[@idkusername2795]

I’m working on an arguing powered 4 legged robot with 3 joints per leg for a school project and it’s my first ever time making anything with coding and electronics. It’s cool and reassuring to see that a lot of what you said are things I used for my robot, my code is probably terrible when compared to professional one, but it will probably still do the job once finished. Figuring out some of the calculations is harder than I expected, but it’s so satisfying once done.

[@DommeDakDuif]

I think it would be logical to have Hyce join on this. And this time you would be telling the stories about your job and how the technical stuff works. After summer break I'm going to a school where I will be learning these sorts of things. so I am very excited for this to become some sort of series. I love the very technical side of scrap mechanic and math so this will be amazing if you make more and explain it :D

[@eddielowe8189]

One of the best videos on robotics I've seen, would love to see more content like this.

[@wazz_up_dog]

I hope for more of this style vid, love the educational aspect and you are great at explaining how this all works.

[@andrewseely5048]

"it was a lot. It was technical" and I loved it

[@logicbuilder1204]

Dude you have to try plasma again now that the full release is out. Stuff like this is exactly what plasma is meant for, and I found myself having a ton of fun making walkers using the exact methods you used here. Super cool stuff!

[@WildZ_25]

Absolutely loved the graphics you did and explained the video. So well done Kan!

[@outandabout259]

This is really cool and similar to what I love doing in scrap mechanic. I like trying to make better and better walkers with number logic controlled legs, and I'm hoping to one day build a walker that can walk in any direction (forward, reverse, strafe and diagonal), turn left and right moving and stationary, raise and lover its height (all of these I have already managed to combine in a six legged walker although the result is a bit janky), walk over uneven terrain while maintaining a constant height without bouncing (requires each leg to know when it is on the ground and and not push further down), keep its legs vertical on a slope, and any number of additional automated features I feel like it would need to be "complete". But it's really hard. Oh, and do all that while controlling legs so that no leg needs to slide for a movement to be possible.

[@TheMostUt]

kAN breaking down Gcode for those who don't deal with robotics, love it. YES tell us your stories kAN! Also, for the video bit, recreate the malfunction, if you can, to illustrate the failure modes. It would be nifty to add pistons in the arms, so that it can extend a bit, and could keep a specific angle range on the main arm elements. Block spawner mod to make a 3axis 3D printer? Or maybe even a Delta design?

[@typhonhunter]

Hey Kan! Love your Videos. Great explaining and math of the robot. Really liked it. One tip with the negative axis thing: u would need to use arctan (probably with quadrant correction: atan2), it returns a unique angle for every position no matter the quadrant.

[@MR_CATGaming]

Love your videos Kan! I would love to hear about your stories from your job in engineering

[@Soken50]

Thank you so much kAN for reigniting some deep trauma from my time in mechatronics engineering school, I've seen enough matrices full of trigonometry for a lifetime.

[@Pystro]

For mods on the end, in the mean time you could just place a spudgun and make a robot that carves art out of cardboard/glass. For the angle D, you can skip the trigonometry and just use the fact that N+C+M=180. Where M is the equivalent of angle N (the in-the-triangle part) at the D joint. Similarly, if you have arms of equal length (so that N=M) you can skip the trigonometry and use that same fact to calculate N+M=2N=180-C.

[@DigitalJedi]

I love these engineering breakdowns. I'd like to do something similar with logic stuff in the future. My PhD is in Processor Architecture Design, and with SM having every basic logic gate, I can build anything.

[@HampusNyholm]

Its kinda fun that you decided to work on robots in scrap mechanic since i made one just before summer. I made a six axis one and i can tell you i got tired of all the connections between math blocks but otherwise it was a really fun problem to solve. Last year i started studying at a university and had a course in linear algebra and that sparked the idea. Regarding the linear movement (which i havent implemented in my robot) i think real robots use the so called inverse jacobian matrix. It is a matrix that correlate linear velocity to joint velocity. But it seemed like a lot of work to implement it in scarp mechanic so i havent gotten to it yet, there may be a more simple solution

[@wihannferreira2721]

As an engineering student now studying mechatronics (robotics), this is really inspiring me to study harder.

[@leoritchie9635]

If you can't get mods to pain/weld stuff, you could make parts that you need to cut out. A spud gun could shoot out cardboard as a demonstration. Not really putting a car together but you could disassemble one. Doors, windshields, wheels, hoods, etc

[@samuelsalazar4935]

This is a nice way to teach math! Seeing the cool things you can do with numbers and basic maths really it´s amazing, and relatively cheap if you compare it to real life XD It would be nice to hear some of the stories you have about robotic too :D

[@poomonyoutube1607]

As a student in robotics, I would very much love to hear some of those stories!

[@Ekipsogel]

When you do the 6 axis, have the end be a vacuum pump and chest with blocks in it, and now you have a manually controlled 3d printer!

[@corbintheintern6229]

It blows my mind that you actually implemented Inverse Kinematics into Scrap Mechanic

[@moonwatch6709]

plsss tell us more stories while doing more robotic stuff, that would be absolutely dope

[@grandevox]

Please continue making robotics series! I used to do a little bit of robotics back in uni (it was my final project) and I thought I was really into it until I find out that I wasn't cause of how complex all of those physical calculations are lol. Figured it's better to just stick to being a software engineer. But these stuffs still seem really cool even though I could only get at most 50% of what you explained, but still it was an interesting 20-minute lecture lol

[@janserjak1191]

Hey kAN i am a huge fan of u and u inspired me to study mechatronics now and i am now finishing the project of building my own 5-axis robot arm at home. Would deff love more videos on this topic.👍👍

[@rajarshibanerjee1533]

Heres how to always get the end bit to move in a straight line. Compare all angles and find the largest angle (if theres no speed limit, take any angle). Find rated speed of motor (same for all hopefully). Let t = time taken for motor to complete largest angle. Lower speed of all other motors to take t time to complete their rotation. So essentially, if all motors take the exact same time from start to end irrespective of the angle they achieve, the end bit will always follow a straight line. Thats the easiest/fastest way to do it without any extra burden on calculations as a single comparison is almost O(1).

[@branceonchristian9606]

This would be really interesting to watch. You get to do what you love, and we get to watch interesting mechanics of engineering and learn math.

[@Tom89194]

I am fairly sure the angular speed for each joint would change as you travel along the vector between two lines, especially if the vector line is long. So maybe the best way is to draw a vector, generate points along the vector (more points means smoother movement). Generate an angular speed for each joint for each step along the line and then probably apply some smoothing function to the angular speed changes....

[@grummanschumann6475]

ive never seen this happen in scrap mechanic, this is so cool

[@wildwyatxbox]

This is very interesting. I'd also like to hear more about the flamethrower incident.

[@johncage5368]

Nice robot for a start. Looking forward to see a 6-axis one with nice linear point to point motion. We definitely need a bar robot that can reliably serve drinks to engineers sitting at a bar without decap ... unfortunate accidents. I love the Mod Pack, makes Scrap Mechanic easily 200% more interesting. A little hint: It would be a lot easier to follow your trigonometry explanations, if you'd follow the usual conventions, and not just name everything that's not quick enough to run with capital Latin letters.

[@ElderonAnalas]

i can't wait for the 6 axis, and the robot controller. and then, perhaps a system for, like, "grab block POS 1, place block at POS 2, grab block at POS 3, place at POS 4" etc, and perhaps a way to get this to automate like, a survival mode factory or something.

[@asimovstarling8806]

Here's a wacky task. Build an 8 axis version of this, then built five more ot if, and then build a board with all the logic and math stuff on board, as well as the arms, and have it walk.

[@carinthesunset1971]

Hey kan! I'm currently studying mechanical engineering and would love to hear any stories you have. I also would like to hear about how your channel started and what pushed you to do KZbin full time instead of sticking with robotics. Love the content!

[@ScrapkAN]

Please upload this to the workshop. This would be so fun to play with!

[@x_540]

I didn't watch you for like a year and then i see you make this banger. All very cool!

[@terror8354]

You could attach a spudgun on the tool head and then put a block of cardboard next to the arm so it could carve out shapes. also please upload to workshop.

[@AshesandEmbers]

Good work. You could attach a spud gun and have it make a 3D picture by shooting blocks away from a stack.

[@unusualfabrication9937]

I absolutely LOVE this. and please tell us your dubious robotics stories

[@ivanljujic4128]

You could add a vacuum pump at the end of the arm so it can build something. in fact, you could have like a cardboard tray onto which the robot places blocks, and that gets destroyed with spudguns and then the robot could build movable creations :D or you could use spudguns on the end of the arm, which could then carve out something from a big brick of cardboard

[@circuitgamer7759]

I would absolutely love to see more videos about this! I'm 15 minutes in, and I want way more than 7 more minutes of it :)

[@England91]

A good half and hour to an hour would be ideal

[@robschilke]

I don’t understand math at all but damn it was cool to see KAN get so passionate in this video.

[@littlebittygames2395]

This isn't quite my fancy but I still love how excited you are about it ,so I watched the video and liked I hope you make more videos like this!