hopping humanoids, continued

kert · 2006-12-17T12:01:15+00:00

I was reading through some earlier correspondence in this group insearch of encoder links and i stumbled upon a "hopping humanoids"threadJohn Nagle was writing this:Why would you want a mechanical solution to energy reuse problem ? Whydoesnt electronics cut it ? I mean, electric vehicles reuse theirbraking energy, and some types are also reusing energy coming fromshocks, i.e. instead of converting that into heat they are dumping theenergy into capacitors and recharge the batteries using that later on.So with robots, why wouldnt you dump the energy into capacitors andreuse it ?With springs, you are converting kinetic energy into potential inspring, and then back to kinetic. With electronic solution you wouldrun your motor ( either linear or rotary, doesnt matter ) as agenerator when "braking" and dump the energy into capacitors,supercapacitors, batteries, whatever fits the bill best, and reuse itto move the actuator again.I know that when using DC servos energy reuse its tricky to accomplish,but most industrial bots are using AC synchronous servos which aresimpler in construction and thus also cheaper, and energy reuse isrelatively easy electronically to accomplish. The only catch is that ACcontrol is more complex algorithmically than DC, but its allimplemented in controller chip silicon anyway.I believe that ASIMO and most of the advanced asian humanoid bots areactually using AC servos, as AC servo has all around bettercharacteristics ( better power per kg, better torque curves, cheaperetc ) so i believe they would go the extra mile and reuse the energy aswell. Call it "electronic spring", the principle remains the same asmechanical spring, its just a matter of different energy conversions.And AC motor to capacitor and back should yield over 80% efficiency.Also, electronic springs are obviously more controllable.further discussion welcome :)-kert

Dec 17, 2006 26 Replies

Rate this thread:

John Nagle 19 years ago

Very nice. I'd noticed a few months ago that the ICs for that were available, but hadn't been integrated into systems. 5DOF is sort of OK, although not having yaw rate means you may have trouble controlling spin (rotation about the vertical axis) while running. But that's a second order problem. You now have enough hardware to do running. And a platform that can survive falling down.

John Nagle

Vote

John Nagle 19 years ago

Not quite. A few have been using single axis rate gyros run through servo mixers to increase standing stability, but not a full IMU.

This is the point at which control moves from hacking to heavy math. The same thing happened in the video game world when graphics went

3D. Early 2D games were all about clever hacks in assembler to make the hardware do interesting things. Then came 3D, and everybody had to learn matrix algebra. Which game programmers did. Robotics hobbyists are going to be spending more time reading math books and theory papers over the next few years.

John Nagle Animats

formatting link

Vote

Joe Strout 19 years ago

Agreed. But there is a standard, widely accepted definition of running, and it's simply that both feet leave the ground for some part of the cycle. Using words in standard ways aids communication.

You really should quit assuming that I haven't. :)

Best,

- Joe

Vote

John Nagle 19 years ago

There have been a few attempts, but they're mechanically clunky, like schemes where a string winds around a pulley with a corner cut, so that when the string is fully wound, a bit more motor rotation pushes the string off the pulley and the string releases all at once. Some people at Stanford tried that. Sort of worked, but not too well.

The best known solutions for larger robots are either hydraulic or pneumatic. A muscle can be thought of as a spring with an adjustable spring constant and an adjustable zero point. You can get that effect with a double-ended pneumatic cylinder - high pressure on both sides makes a stiff muscle, low pressure makes a limp one. Unequal pressure changes the zero point. That's been tried, too. Works fine, but you need to carry an air compressor or have external hoses.

The same trick can be played with hydraulic cylinders and accumulators, and that's been used in Boston Robotics' Big Dog, which is pony-sized.

For desktop robots, high end servos, maybe with some stiff springyness in the legs to absorb shock loads, seems to be the way to go for now. Right now, we need better control algorithms, not higher efficiency. Energy recovery will improve battery life, but that's not the current top problem.

John Nagle Animats

Vote

kert 19 years ago

What about pair of magnetic clutches to disconnect the servo transmission and hook it up as a generator ? Sounds clunky and expensive .. Also, what about reversible tranmission types like planetary or harmonic drive ?

-kert

Vote

dan michaels 19 years ago

So hopping is running, then.

You might look up the differences in gait biomechanics, vis-aviz foot action and energy untilization. How does the foot pattern differ between walk/trot and gallop. Why does a transition occur. Why doesn't an animal gallop at slow speed. Why can't an animal trot as fast as it can gallop. Why does the inverted-pendulum apply better in one case, while the spring-mass model is better for the other.

Vote