LIMS, the Fast and Safe Robot Arm - Elbow Mechansim

[@vmened]

State of art in robotics. Bravo!

[@joannot6706]

What is special about this joint is that the hands joints doesn't require the motor to compensate for the elbow joint movement in a design using cables. That's the reason behind that whole look. The only way that they found (and that I found yet) to do this trick is to have 2 perfectly parallel rotating axis on the elbow.

[@TheMarcusrobbins]

why not compensate in software?

[@zephgaddielp.sanchez8914]

True true....

[@calllen]

Is there a name for this kind of joint? Like a hinge where the two parts roll around each other rather than both parts rotating around an axis.

[@IRIMLabKoreatech]

Usually, we call them the rolling contact joints. These have a long history, please refer to patents.google.com/patent/US3945053

[@oompalumpus699]

Damn it. The quiet background and the guy talking in a low voice made me think I was watching JAV.

[@epicn]

Did not expect to see a comment like this here

[@markcollins2538]

NIce! What material is best for the wire?

[@muxammadrizo3643]

can you tell which CAD you used to simulate dynamics?

[@cayenne7792]

why is it fast? why is it safe? I see a lot of pinch points....

[@LOC-Ness]

like a bicycle... interesting

[@tuoananh4559]

Where can the arm joints you buy me tell me?

[@Kimchi_Studios]

짱이네

[@korso7717]

Fr, iam doing this too, i really like it but it's useless, expensive and the arm is not strong enough, but it's still a smart invention

[@AlxM96]

Useless? It has great benefits such as a torque amplification factor of 6 and a stiffness amplification factor of 36, it has very low losses due to friction and it's relatively simple since it uses few parts compared to other alternatives commonly used in 7DoF (degrees of freedom) robotic arms. As a matter of fact, it has been developed up to a commercial product, currently marketed as the NAVER Labs Ambidex

[@ulforcemegamon3094]

@@AlxM96 also i highly doubt is expensive since is mostly just a bunch of wires , compare a bunch of wires vs having to buy many motors , and such motors are expensive too

[@AlxM96]

@@ulforcemegamon3094 you're absolutely right about the fact that the cable-driven design indeed reduces the cost, due to the fact the motor is not mounted *at the joint* (rather at the end of the arm as a counterweight) its own weight doesn't have to be lifted together with the arm itself, allowing for higher payloads and permitting the use of cheaper, possibly heavier, and less power-dense motors to be bought. However there's no way around the necessity to use _n_ motors for _n_ degrees of freedom (dof), if you need 7 dof then you need 7 motors. The trade-off is in the complexity, construction and tensioning. With traditional designs, you can just mount a motor on a linkage bar and then repeat, displacing the inertia of the motors is the real challenge. I am currently designing a low-cost cable-driven robotic arm myself and it's not trivial at all.

[@ulforcemegamon3094]

@@AlxM96 yup , that is why the whole arm (not including the shoulder , that one weights like 7,5 kg) only had a weight of 2,1 kilograms and since the weight is not that much and is distributed trough the arm , it gave the arm an very low inertia , which enables it to be quite fast but also being able to easily stop even at fast speeds , you are right , this one is quite complex regarding the construction and tensioning , so i hope you can design that robotic arm without too much trouble

[@AlxM96]

@@ulforcemegamon3094 Ah, a fellow paper reader :) and thanks! But I unfortunately forsee mine to be even more complex ahahah it's a necessary compromise, we want to locate all 7 motors (cheap but bulky) in the "chest", and route all the driving cables through the joints they cross (i.e. wrist has to cross through shoulder and elbow). I'll be posting on my channel when we begin testing so hopefully soon!

[@rohitroll2119]

Just get a hinge. What's the big deal?

[@AntiNoobDefence]

This one has better degrees of freedom

[@thegreatzoom5073]

Not only that, but it seems extremely precise. Just a hinge won't do for the applications they have in mind.

[@cyberlord64]

@@thegreatzoom5073 what do you mean by "extremely precise"? Precision is determined by the actuator, no?

[@thegreatzoom5073]

@@cyberlord64 Precise in its arc of motion, not necessarily how precise the angle it can hold is. As it rotates, there's almost no give on the radial axis. Just as an example, if someone used a door hinge for this it would have a lot of "play", while this design eliminates it entirely.

[@joannot6706]

What is special about this joint is that the hands joints doesn't require the motor to compensate for the elbow joint movement in a design using cables. That's the reason behind that whole look. The only way that they found (and that I found yet) to do this trick is to have 2 perfectly parallel rotating axis on the elbow.