These ARE The Droids ...

[cmay...@gmail.com]

Dan Albert and Chris Mayer are starting up the second wave of building a full sized humanoid android.

We've done it before, and want to take what we've learned from our first sets of walking robots to the next level.

(Our temporary code name is "These ARE the droids you're looking for.")

(WATSON on the table, May 2016)  https://youtu.be/ChYZN1z0_mc?t=821

(Walking demo, July 2017) https://www.youtube.com/watch?v=BLBBoPJLEh0

Join in on the fun, especially if you could help with the following.

    Mechanical Engineering

    Robot Simulation

    Software Engineering

    Electrical Engineering

    Power Distribution

    3 Phase DC motor control

    3D printing

    CNC milling

(We're looking for active members, not arm-chair quarterbacks.)

We plan on crowdfunding once we get close to a working prototype.

If you might be interested in joining this project, please reply privately.

We're planning a kick off Zoom this Monday evening.

[Chris Albertson]

I also built a walking robot, it was a dog-bot.  I learned a lot.  One thing is that walking robots are like airplanes, where weight and power really matter.

OK, so I assume the photo is of the old robot and you are starting over.

One big challenge will be just getting the process in place.  How to keep requirements docs, CAD, simulation, and the real robot and testing in sync when the project is distributed.

About three-phase BLDC motor control.  There are two parts to this (1) software and (2) hardware, and it will need to be custom.  For software, just use this: https://simplefoc.com/

It is completely open source and is easy to use and runs on a dozen different platforms, from ESP32 to Pi Pico.   Small prototypes can use breakout boards for hardware, but in the end, you will need full custom PCBs. SimpleFOC allows you to control position, rotation rates, and torque, and the community is very good, with many experts there.  I have a few motors in test stands here.   (fun fact:  software bugs can be expensive when you use 80 amp power supplies.)

Choosing a simulation environment is going to be important.  You want a realistic physics engine.  I tried some cheap and low-learning-curve options, don’t bother.

The problem I found when trying to build a second robot dog that could trot and jump was the cost of the motors.   For the power I needed, it would be hard to go under $200 per joint.  A full-size humanoid will require some quite large motors if the goal is walking outside of the range of static stability.    I doubt you can build for less than Unitree's claimed $5,600 price.   Getting it that low would be hard.  You need strength, but more than that, you need high control bandwidth (high accelerations over small angles)

Finally, one question:    What are the performance goals?   Should it be able to run outdoors or only do an indoor shuffle-walk?  The answer affects the cost in a big way. (It reminds me of the "rocket equation")  A shuffle walk can be done with servo motors, a run needs some very powerful BLDC motors, maybe “borrowed” from industrial-size drones.   Then there must be a payload goal for the hands.    Will it walk up and down stairs?  What about uneven hiking trails with possibly loose dirt?    Could it ride a bicycle or juggle three tennis balls?  And what is “full size”.  Many adults are less than 5 feet tall and some are 7 feet.

You said you were looking for CNC expertise.   Mine is minimal. It is going together slowly.  Today it can draw with a sharpie pen, but I’ve nearly lost interest because today we can send files to China and have anything made for cheap.   My calculations show me that you can make a large robot like this using plastic, but you will 100% need many metal parts as plastic needs to be physically large to have the needed strength, and then your part can’t fit within the outline of the robot.     The way to go is “stressed skin” and not a hard internal frame.    The best example is a good battery-powered tool like a Dewalt drill.   Mostly a plastic structure with metal stress and wear points.   You can print most but not all of it in “engineering plastics”.  

Please keep the discussion out in the public where people can see it.   If it goes well, people will see you have some chance of success.   This is the hard part of getting an open source project going, showing potential contributors that it will work.  A lot is organization. Flowing requirements into design and simulation and test, and then manufacturing is not easy.  GitHub can be a big help as it can do version control, forks and manage  problem reports

BTW the name “These are the droids.. “  is good.  

--
You received this message because you are subscribed to the Google Groups "HomeBrew Robotics Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hbrobotics+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/hbrobotics/873000e6-9b81-4b61-b153-b564c5f19e87n%40googlegroups.com.
<20160818_180619.jpg>