Thank you for the kind words of encouragment. Everyone has been so positive.
Milo’s tracks are made from REXNORD roller chain for conveyor belts. They are schedule 60, stainless steel, and weigh about 13 million pounds. (that’s what they feel like when you drop a segment on your foot)
I got short segments cheap on eBay. I then glued rubber pads to the stainless steel plate for traction.
I have posted additional photos which I hope will help you to understand how I made them.
What A huge robot ! how much total weight and height of this robot ?
==========================
quote: "For the arms to be detachable, all the motors that drive the shoulder lift, elbow lift, wrist rotation and gripper open should be on the arm itself"
Brilliant work, Rebel_Ace… I’ve got to say I admire your attention to detail in every aspect of production. Do you have any plans on releasing the .NET source code at all?
As mentioned above, MILO stands about 5 feet, 4 inches, and weighs about 300 pounds.
The motors MILO uses are large electric motors that are surplus items from either the automotive market or the electric scooter market or garage door lift market. In the case of the actions of Shoulder Lift, Elbow Bending and Gripper Open/Close, the motors are NOT where the joints are. The mechanical force is conveyed by steel cables. I showed photos in various stages of assembly, so you can’t see the cables or motors in some cases. Since the arms are directly taken from the old “Scout Bot” build, you can watch this video of the arms being tested under manual control. Video of Arm Movement
Thanks for the compliments. Taken as a whole, my source code is something of a psychotic mess. As far as releasing any code, I would probably show individual routines or snippets that do something worthwhile or interesting. That way, for folks coding in other platforms, they would not be overwhelmed by all the Dot Net overhead coding that might not be immediately germaine to the task being done. At some point I was going to write something up about the “Face Detection / Face Tracking” routines. Since I just ordered a baker’s dozen sweet new rotational encoders, I will be doing some mechanical re-work, and re-writing some of the motion control routines to take advantage of the capabilities of the new encoders. When I am done with that, perhaps I’ll show some of the code for review and critique.
I read about that very nice hardware build and was wondering if the programming was on the same level… … and then you blew me away with a really nice remote console and even a interactive 3D model! ^^
Thank you for the kind words. I checked out your Ping Pong ball “juggler”, and was very impressed with its “evolution”.
In particular, I was completely geeked over the iteration where you experimented with sound to locate the position of the ball. Outstanding, and something I probably would never have considered. Very, very clever.
Just signed up to this website and yours was the first project I came across. Needless to say if this is the standard of other project on this site, I have much to catch up (just a beginner).
Thank you for such detailed description of the work and the effort that went into making this robot.
I am sure to be interested in more updates that come from your side. All the very best for future projects too. Once again a very professionally executed project and it deserves a pat on your back.
Trully an amazing piece of engineering. My hats off to you for the amount of dedication and detail and devotion you have put into this project. You my sir deserve a pat on the back and a steak dinner. Please put a video up when you get a chance. Great job you did.
Too bad about everything that could go wrong did go wrong while attending the science show, still the interaction between Milo and the kids was nice to see.
How are the threads holding? There must be a whole lot of force pushing a pulling on those glued rubber pads.
Since the treads were first used on the ScoutBot, that makes them about 4 years old. They are in surprisingly good shape. I have just one that has partially separated from the stainless steel plate. When the robot is moving in a straight line, there is very little stress on the pads, with respect to their plates. However, when the robot is turning, the differential drive produces very large shear forces on the pads, but those forces work across the entire surface area of the glued joint. They DO NOT want to slide off.
However, the pads are subject to “peeling” off, if a foreign object manages to wedge itself between the plate and the tread, which is what happened on my partially separated track.