Это видео недоступно.
Сожалеем об этом.
DIY Doc Ock Suit Pt. 1 | Designing a Collapsible Arm Mechanism
HTML-код
- Опубликовано: 13 июн 2022
- Use code MRVOLT at curiositystrea... to sign up for just $14.99 for the whole YEAR
Support the Ukrainian people: crisisrelief.u...
Support the channel: / mr_volt
Follow me on Instagram: @mr.v0lt and together we can finally take down Spider-man.
The man, the myth, the legend: Thang010146 • Sarrus linkage 1
The MathMechs extensor toy: www.mathmechs.com/
Sarrus linkage credit: en.wikipedia.o...
Use code MRVOLT at curiositystream.com/MRVOLT to sign up for just- $14.99 for the whole YEAR.
Subscribe to help me -destroy- hug Spider-man with my harmless cosplay.
hug him so hard the rest of his suit turns red
You really calling us out with that ad lol
The problem is that it not only has to be able to extend, but that extension mechanism needs to be able to bend in any direction and at any extension amount.
You can figure it out, I believe in you!
and i think he needs to also make at least some segments so it can make an S, not just a simple bend, by compressing somehow one part of the mechanism
Loved the Wheatley, can’t wait to see the Ock arms!
The mechanism reminds me of those extend-o-arms that work with a scissor motion.
Being flaccid is good for a continuum arm. Drive a series of tendons through the mechanism to point the arm in different directions like doctor octopus 🐙
Worth a try! I have to keep reminding myself that it doesn't have to be precise motion, but just needs to look cool and move around haha.
I did kind of build something similar to this but it was broken in couple seconds and it needed way more energy than I could have gave it
Looks like you need to hold the power of the sun in the palm of your hand.
@@tvbot6496 lmao
Steve Johnson's XFX did the original pieces and assemblies for the movie. It was mostly rod puppetry with the finer elements being cable driven. I always thought of doing a cable version, but subdivide the stages with each having it's own cable control apparatus (would make the math easier).
Would need quite a few cables! I've watched the BTS of Spiderman 2 so much now too. Ah, yeah it's wild how complicated it is as a puppet. I'm not attempting something at that scale, but if it can articulate and be operable by the user, I'd be happy.
This is certainly gonna take a while, but failure is only a part of the process until eventually you get a bright idea! Can't wait for more dock ock vids.
The Spider-man 2 Otto Octavius arms are reason why I start love mechanic inventions when i was young so its good to see you as one of the coolest inventors tried to make it. Good luck!
Thank you!
If you want to be able to control the bend of the arm and get it closer to prehensile, I feel like being flaccid is a good thing. Theoretically, you could have another mechanism - maybe threaded cord through each of the three corners? - that stabilizes and controls the bend of the arm to reach around things
Perhaps the "flaccidity" there at the end could actually be a bonus. use it to control the curvature of the tentacle with three cables attached from the end to three motors?
Yeah, I was thinking the same thing. It it was rigid, then the arms would just stick out straight from the back, which would look kind of ridiculous XD
Though the other problem there would be that (I think) the 3-cable design kind of needs a more solid core, otherwise the pulling cables would just make it compress more rather than bend all that much
09:26 You went there, I hoped against hope you would, and you didn't let me down. I'm proud of you 😂
Looking forward to seeing the problem solving dude!
hell yessss new Mr Volt upload!! love ur content man keep it up :D
Yes his content is really great
next episode: "I AM BUILDING A WORKING DEATH STAR"
Huh
My brain hurts just looking at the mechanics - looking forward to see how this project progresses!
Please continue this project!
I found you via your Portal content but I’m also a big fan of Doc Oc!
It has been my dream to create a function version of doc ock's arms. Since I don't have the resources to do it myself, I eagerly await the progress you make with this!
I did start a version of this once, a continuum robot that was cable-driven rather than having any hinges. I got it all working in theory except I couldn't reliably attach the cables from the arm to the motors to drive them (and then subsequently abandoned the project).
Don't give up. I made a set of working arms for less than $1000. My biggest problem eight now is coding. They work pretty well mechanically but I can't seem to figure out the code.
Dang I thought I recognized that style of thumbnail, amazing to see Thang010146 finally get some good recognition
You can use the flexibility of the final version in the video and use wire connected to servos to control the arms
Yes! Glad someone else has mentioned Thang's channel. So many excellent mechanisms in his videos. They've been pretty invaluable to my projects, as well. They're also a great thing to browse through for inspiration.
The way i would tackle this problem is with a structure that is naturally springy with it's resting state being extended and a cable and pulley system to pull it in. This would allow you to more properly solve the bending problem in it's resting state
I think that's the way to go. Keeping it preloaded at all of the joints will be easier and a faster extension action. I'm not to concerned with retraction speed, so a ratcheting mechanism could handle that.
I love Thangs video so darn much, they have been the start of so many projects.
Glad Thang is getting a shoutout, been a fan of his for 6 years. :)
ooooo this new project is very exiting im patiently optimistic for what you come up with next
I have thought about starting a project like this, right now i don't have the money or time. BUT my original idea is to use a "snake-arm robot" have you thought about a design like that? It doesn't have to be complicated it could be controlled by 3 cables and 3 linear actuators. If i ever get the time to build one I'll share whatever i came up with. We all love doc oc
I've looked into it! I may have to ditch the extension/contraction element I'm playing around with here. My only issue is snake/continuum robots are even less stiff at small scales.
I'm excited to see where this goes
I just discovered your channel and I am in love.
I think you may have to think out side the box for this when designing tentacles. My intuition would be looking towards tubes of water and ball bearings to get the fluid motion of tentacles.
Oh! I just checked. Can’t wait for part 2!
Made me think about newer research in inflatable tubes that become ridged with pressure. If you had three tubes (like the member you eluded to) running down the center I bet you’d gain rigidity and prehensile capabilities 😸🤷♂️
fantastic work sir! Love your content
That's actually what they did in into the spider verse!
ooooo! I'm excited to see your process/progress on these! Doc Ock's arms are SUPER cool!
Have you ever professionally worked with animatronics for companies?
No, but that certainly would be fun!
Use the the non rigidity to use string drives. Then you can control the rotation movement at the end
I agree with the comments of adding cables to add rigidity.
time it out so the arm extends at the same rate the cables release. the cables keep the arm rigid in compression while the lead screw and arms push out in tension.
or even better, do three sections of three cables. then you can have each arm basically have three individually articulated segments.
If there’s one thing I’d love to see you tackle, it’s either the Dinergate or the Manticore from Girls Frontline. One is a robo dog with a gun for a tail, the other is a massive, 4-legged armored tank.
Hadn't heard of this before! Just looking at some still images, the designs seem "plausible," just highly stylized quadrupeds.
Have you looked into soft robotics at all? That might help with the articulation you’ll need to actually move the arms
I have! Lot's of cool research being done. The biggest issue I have is having a portable tank and compressor makes portability an issue for something that needs to be a backpack.
This is so cool! I'm so excited for more!
ah nice! i HAVE seen the great thang before! you probably came across it in your travels, but i've also seen a very cool, very simple soft robotics example that you might find useful. from what i remember its basically a hose folded into itself. scratch that, i decided to try to find it and it was really easy.
'this unstoppable robot could save your life' by veritasium
Doc ock arms but you have to blow them up like spongebob's anchor arms
Love that video :) It came up while I was researching. I spent some time planning and thinking through a pneumatic version, but rigidity seemed like a major issue. They only really show one variant that can hold its own weight for a rapid antenna mast. I think it's still worth investigating.
@@MrVolt I think the flexibility on the version you showed at the end of this video could actually be a feature - you could run 3 cables down each corner of the arm, attached to 3 motors at the base. the leadscrew would provide the extending force while the 3 cables would tension the entire assembly up, as well as take advantage of the backlash in the links, allowing directional control by selectively extending the cables.
@@KineticWasEpicVideos the sadly seemingly retired channel 'geared down for what?' made an arm like this at one point, with degrees of freedom way beyond 3. i believe its also the approach jlazer took for his doc costume. and i think it was festo that made something similar on a more or less professional level too. they're all ok but were all much to compliant to control with any precision or apply much force.
same obviously goes for the soft robotics i suggested too.
its a tough tough design challenge, i'm pretty stumped for ideas.
i think the best i can come up with are actuated ball segments a bit like the clickbot or big hero 6 thing, assembled like the helping hands adam savage likes so much. but would be many many many individual motors, no retraction, and and and.
you have fantastic ideas though and your challenges always really inspire me to get creative, so very keen to see where this goes @Mr.Volt!
I really don’t think linkages are the way to go. Less moving parts is better.
That’s why I would suggest going for something pneumatic instead.
I understand the want for having collapsible arms, but these mechanisms ideally only allow extension in one linear axis, and not the complex movement of an actual tentacle.
And this is inherently the issue with Doc Ock, or similar bio-mechanical suits, they're just not practical to control, at all. Just look at something like Festo's various biomimicry stuff, specifically the elephant trunk (Bionic Handling Assistant, BionicSoftArm, Bionic E-Trunk) and look at how that's actuated; the absolute simplest design is having three connections between the 'vertebrae' to actuate rotational movement around the vertebral pivots, though I believe there's also designs that use four and a pulley actuator for opposing sides for two motors rather than three, there's also fully pneumatic designs. Having a corrugated spine would make this collapsible, as you could pull the cords or collapse the pneumatic bladders to compress the arm entirely, but this eliminates the option for four wires with a two sided pulley drive, which returns the mechanism to three actuators per section. But this is the issue, it's a number of actuators _per section,_ in a small system you're going to have at least three sections if not more realistically four or five, in a full scale limb like Doc Ock, you're probably looking at ten or more just for coarse control, ideally you only want a handful of vertebrae within a section for ideal maneuverability of the system. The first of this issue is control system, as you have a ton of actuators, you're going to have a complex control scheme no matter how you set it up. The second of this issue is how you activate that control scheme, even if you have preprogrammed movements how do you plan on actually activating those movements.
One model is ideally rigid, though flaccid and potentially compliant in practice, while the other model is overly complex in design and control, and there's really nothing between the two that's nearly as practical. I'm curious to see where this project goes, but realistically you'd be better off developing an arm and developing an autonomous robot off of that, as there's just no realistic way to control extra limbs, even as a prop.
I appreciate the thoroughness of your reply. I didn't present a project overview in this video so it's led to some confusion. You're right when it comes to the level of articulation necessary to have coarse control. It's simple not feasible. Practically, I'm trying to limit each arm segment to only have 4 DoF maximum. Two at the "base" to allow rotation and a single curling action, this extensor would be the third, with the fourth yet to be decided. I'm keeping a level head about this. I've spent a fair amount of time researching continuum/snake robots and the number of actuators has to be kept to a minimum since I'll need 4x for the full "suit." And the kinematics of anything beyond that is far beyond my skill level. For now I have plans for a fairly simple analog-based control that will be attached to my actual arms. Simultaneous control of four separate arms isn't in the works, but if I can effectively switch between two arm control fast enough, that will satisfy my goals.
@@MrVolt Realistically what would the fourth degree of freedom be? The third rotational freedom, abduction and adduction, is irrelevant due to mixing of what's already available, circumduction and flexion, and the two other lateral movements along the X and Y axes, assuming Z is extension, are also irrelevant to the joints as you don't need planar movement; 3dof, rotation, flexion, and extension, are really all that's needed per segment. But in the models I pointed out, the ones with three actuators per segment, can induce a form of rotation per segment due to mixed flexion between the actuators, removing the need for a discrete rotational platform per segment, and the controls for all of this can be localized within the backpack, as pneumatic tubing or wire (probably preferably within a bowden tube setup, though I haven't seen a collapsible bowden tube) don't need their motors at the segment; the benefit of this is that extension would also be a mixed input of just pulling all wires taut, or collapsing the pneumatic bladders, assuming you have a spring within the spine to return the segments to a normal position (normally extended for wires, normally collapsed for pneumatic bladders). Kinematically more complex, though I'm sure models exist somewhere, but mechanically a lot less complex for a long arm; and the kinematic complexity is entirely arguable, using a rigid arm with a base spherical or faux-spherical pivot with an additional extension degree of freedom could be just as complex kinematically, could even be more complex, I haven't really seen a 'tentacle' produced with rigid arm segments before. The only joints that would really require a fourth degree of freedom, of which being a true rotational platform, would be the base of the arm, if the base even truly needs this, and the end effector, of which the end effector could also be driven by wire; I believe Skyentific may have a video with a wire driven pivot within one of his robotic arms, or maybe one of the LIMS2-AMBIDEX from IRM had this kind of pivot, though there's also been various other projects using wire driven limbs and joints that may have used a wrist pivot, it's been quite a while since I've looked into these projects so my memory on them is a bit hazy.
Personally, I believe this is the way to go, as having a pivot at the end of each section with an extension mechanism won't create nice curvature in the limbs, and would rather be linear sections with a hard pivot point, think segmented spider legs rather than a tentacle. Doing a spherical pivot, or faux-spherical pivot, like this would also need to be more robust, all things considering, and the arm segments themselves would need to be more robust due to the extra weight they'll have to carry for subsequent segments. This is why the Festo elephant trunk, more specifically the Bionic E-Trunk but modified to be a motor-based wire pulling system (the actual Bionic E-Trunk uses a shape memory alloy to shorten wire segments, rather than pull wires with a motor), is honestly genius, as the system is lightweight compared to a more typical robotic arm, and when you have tentacles that're at or above human scale you need a system that can hold itself while not being too heavy to carry. Weight and robustness is going to be a massive constraint for this project, the more complexity you have, the more robust an individual mechanism will have to be, and that adds more weight, it's a snowballing issue as it affects every segment in the chain, times four, within the package of a backpack.
Wire driven segments, three actuators per segment, flexion and 'rotation' are possible via mixed inputs, pulling all three wires would collapse the segments, just have a corrugated (collapsible) spine with a spring that allows channels for the driven wires via collapsible tubes. This leaves a true rotation pivot for only the base and end effector, or hand, of which the end effector can also be driven by wire. Low weight, more robust by design, though potentially more kinematically complex, but also more realistic, and potentially lower cost. Once you figure out the collapsible spine and collapsible bowden tube issue and how to route them, kinematics would be the next issue; of which would probably be simplified by 'training' a program various positions and letting an algorithm figure out all the in-between points, as these limbs would curl rather than bend at a joint, so endstops may be an issue, especially as the opposite side of flexion has to relax to allow the other side to be pulled taut, and this is further more complex due to the 120° offset of each wire within the segment, etc. Probably the most effective way to tackle this project.
This project is already complex enough as it is no matter what actuator design you go for, I just believe that this actuator design is more ideal all things considering. This would be a daunting project even for people who have messed with tentacle-like robotic arms before, mostly because of the scale of the project, usually practical models are used as pick and place robots at a much smaller scale or the implementation isn't under such tight constraints, and the whole extension concept is mostly untreated territory outside of theoretical models, like using a collapsible spine with an assistive spring and how the wires would need to be routed through the robot like having collapsible bowden tubes within the spine, etc. I just think rigid arms at this scale, especially under the constraints of this project, will hit a dead end very quickly, they'll become too heavy, they'll become too mechanically complex, etc. A project like this requires an insane amount of foresight to avoid it becoming a time sink and money sink, even if good planning is involved.
I'm still curious where this project goes and how you execute it. I just wanted to provide some insight and opinions, because I'm sure everyone here is familiar with how out-of-hand large projects can get, especially when there's not many operating models already existing for what the project seeks to accomplish; and I'm sure we're all familiar with how out-of-hand projects and quite easily become multiple year investments, if not also fairly large financial investments. It's an interesting project, just a complex and daunting one, though I'm sure you'll figure out something that works well in the end.
I'm also typing this on no sleep, so I'm sorry if my brain vomit doesn't make much sense, I've tried to make it understandable to the best of my current abilities though.
Thank you for sharing your design process and troubleshooting. Truly inspiring and motivating.
The government needs this man to make weaponary turrets.
you mentioned Thang010146 but forgot to mention Henry Segerman, wich I can see the direct inspiration in one of his new projects. I don't know if it's conhecidental, but your mechanism looks just like the grabber mechanism he has in his channel.
Ah, I didnt know his name. I do mention MathMechs, and have had the link to the toy in the description
ALL HAIL THANG!
If you haven't looked at it yet, a modification of Henry Segerman's Extensor system might be a good alternative if only the center pivoted design might be simpler to drive.
I haven't, so I'll have to check it out. thanks!
Are you planning on lifting things up with those arms? Or specifically expansion/retraction? Maybe it would be easier to start with a couple of linked rods of known length and cover the mechanism with something that resembles the arm doc oc has... Or maybe even having linked pistons, but that would be a nightmare to control later down the line
“Which may be my most ambitious project yet”
GLaDOS Alexa: Am I a Joke To You?
I just thought of a solution made of paper, string, and a tube,
Paper or letter paper would be the triangles, string would control it, and tubes can extend, retract, and can bend, put them all together, doc ock!
I would reccomend working with the into the spider verse design which goes off air pressure!
Sir One question. What did you study? The things you do are just amazing!
Thanks! I dont have a degree, but Ive been doing self-study and had some engineering jobs for about 8 years now. I just try to learn new things constantly, and push myself to design at the edge of my comfort zone when I can.
@@MrVolt Thank You! im not sure what i will study but i'm sure there i want to be so creative like you sir.
I expected this from the beginning, but I wonder could you stabilize the arm with string going throught the middle of those hubs and after extending you wouldpull the string back, but maybe it wouldn't work in reality because of the string being not rigit or links not being strong enought, etc, etc.
you look like if someone combined Derrick from Veritasium with Alec from technology connections and threw in a first year engineering professor
Idk if i'm right, but as far as i know, you could just add 3 strings on the outside at the start and perhaps also 1 string in the middle at the end in this prototype. It should make it less jiggly and it can now turn in theory. The only hard part of this idea is the coding i guess.
anyways, enough talking, give me the Tentaphyl.
Yeah, tendons are definitely in order! I'm more trying to balance weight and cost, since every motor/component has to be multiplied x4 so it stacks up really quick.
nice! cant wait to see the next ones ;-)
Wait wouldnt it being bendy be helpful because then you could add the string as wires that can keep it in place and make it move around
YES!
I'm not an engineer, at all, but would having a rod in the middle of the arm, which extended as the arm extended, add to its support?
Also, why attach the motor to the bolt rather than the nut? Maybe its for fine-motor control or safety, but it just seemed really slow on video.
Could you try spring locks?
Have you thought about those "arms" that medical robots have? They seem to be very movable but I don't know if there is anything about that technology online.
but if they extent like this, how will they be able to move
This is only one element of each arm. The components that allow other motion will be closer to the back.
The power of the sun in the palm of my hand
can we reduce buckling in this when given to a long distance.....
Report for bullying. This guy brags about his tentacle increasing to more than 7 times the original size when extending. Very rude
Awesome! What program are you using?
Omfg the cut at "tentacular dysfun-"
tentaphyl got me good
After you finish the doc oc claws I want you to make a spider man and fight it
I would love to see a samus aran suit from metroid
I made something like this outta cardboard when I was younger, kinda having a fanboy moment right now.
i would have tried a radial way analog to tube sleeves.
Do you have an example? I'm not sure I follow, but am curious.
@@MrVolt certain braided metal sleeves used to protect tap/faucet lines are directional and expand if you twist them. their diameter stays constant due to the way they intervene but the space between the braids expands when twisted in one direction and contracts in the other. but probably a design like your linkages just mounted in a hexagon or triangular pattern could achieve a similar "expendable sleeve" but leaving room in the inside for pneumatic or other forms of axis control as well as for cabling.
+1 for Tentaphyll
It's the good stuff
Can I buy your GLADOS plez ??
Wow I watching this video only 9mins later than it's loaded to utube
Sorry 4 my English
Your English is good buddy
@@Cartzeoon Thx
Can you try to make a tripod from mental gear rising reveanguen?
If I may suggest , inflatable robots might be what your looking for
Inflatable robots are pretty powerful, but the biggest issue I have is having a portable air tank and compressor makes portability an issue for something that needs to be a backpack.
What cad do you use?
This was pretty awesome! I wonder what you could do with Hacksmith Industries money lol
A lot more for sure! Mostly Id use it build a cool team :)
@@MrVolt An awesome team of Robots, Supersuits and CRT monitors lol
if this route ends up being too brittle i'd be fascinated to see you take a stab at something like the tentacle design from the post-credits of amazing spider-man 2.
ruclips.net/video/5CRVxy1omc8/видео.html
(i have zero engineering experience or knowledge)
it's unclear to me if just it uses rigid segments that rotate or if there are also elbow type hinges. I think a version with short macaroni-shaped vertebrae that spin individually could be more poseable and dynamic while maybe being more rigid, sturdy /capable of carrying more weight. it could also still somewhat expand/contract by coiling.
it'd also be cool to drive the arm motion with vr controllers, maybe just a few key poses based on hand position and/or the ability to grip.
I have no idea what i'm talking about but i really hope you're continuing with this project.
Wait who are you again, wait.
OHHHH you're the guy who made glados and Wheatley. Hey how've you been
Ey yo, it's been little over a year, any update by chance?
It’s been so long
Orbital probe canon
i remember a company called mobot. they made remote controlled robots for use in dangerous environments. some of their models had long spine/tentacle like arms that used cable tension to move them. hughes aircraft was their parent company. there are some pics/footage out there
So your basicly makeing a cooler dollar store grabby hand
Yes. I wish it were that cheap though
I’m back remember?
H