Let's Make Robots!

True Human

Just a thought project (as I never seem to have any time to build the ideas floating around in my head);

I see a lot of wonderful projects built here on LMR, but one thing that always screams to me with humanoid robotic designs is 'x' axis is always considered, but the 'y' axis on robot joints is almost always ignored. Please don't take this as critism, rather a challenge to think about this problem.

Think about the human fingers as you type on the keyboard; they move away and closer to you (x axis), but they also move left and right across the keys (y axis). If this 'y' axis movement is ignored, then the whole arm must move left or right to move the fingers into position.

Now, not trying to blow my own horn here, but I have come up with a test 'ping pong joint' that would hopefully solve this issue. As I mentioned, I don't have time to build more of this concept myself, but would love to see if anyone could take this idea further.

Would love to discuss thoughts about this.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Well, here's the thing. The human body is very complicated. And not only the human body, take any body you wish. Our robots mimic bugs, animals and some, even humans. Some crazy talented people managed to create robots extremely close to the real thing, but that is just out of reach for the ordinary robot builder. I think that with the addition of the 3D printers, we might be able to get even closer to the real thing. However, even if we manage to create all the mechanisms necessary, programming all those motors to actuate them will be the road block for us. The programming brings life to our robots, otherwise they are all just RC dummies. It takes an army of scientists to develop a robotic hand to through and catch balls, to pick up eggs or to swing a bat. We are not scientists, we are ordinary people passionate about robots that mimic life. We are always trying to improve our robots and in general we more often succed mechanically and less programmatically. I've seen hexapod robots move life like and they are using inverse kinematics to move like that. I think that a hand also needs inverse kinematics to move fluently and that is not simple to do. Balancing robots use the Kalman filter, another heavy math thing that takes enormous time to adjust right for each particular robot configuration. So, in (my) conclusion:

That is why we simplify things, not because we can't build them, but because we can't program them properly and it will just suck.