Self balancer does it differently than we’re used to seeing

This self balancing robot still uses just two wheels, but it’s balancing very differently than we’re used to seeing. Where most of the projects use a form factor that’s similar to a Segway, this works just like a bicycle. But it doesn’t need to keep the front and rear wheels spinning to stay upright. In fact, the video after the break shows it balancing perfectly while at a complete standstill. [Aoki2001’s] creation isn’t stuck in one place. He included distance sensors on the front and back which are used to move the bike as if by repulsion.

The large wheel where the rider would be is what makes sure the vehicle doesn’t topple over. It acts as an inverted pendulum, pushing against the large wheel’s inertia by rotating the motor to which it is attached. The same concept was seen back in march on a full-sized bike. But why use two wheels when you only need one? His unicycle version can also be seen embedded after the break.

It’s worth looking at [Aoki’s] other YouTube offerings too. He’s got a small robot which balances on top of a ball. It’s the desk-sized version of this hack.

19 thoughts on “Self balancer does it differently than we’re used to seeing”

1. Iw2 says:

I feel like the first sentence should rather read “This self balancing robot still touches the ground with just two wheels, […]”. As you said yourself, it’s using more than two wheels ;)

2. zuul says:

Wow, that works really well

3. jcroll says:

research can make amazing things….

but i want to put an evil knievel doll on it

4. Tom says:

Brilliant.

5. Hirudinea says:

This is now my favourite non pornography Japanese internet video. Physics question, does the flywheel need to be so big, scaled up it would be kind of awkward.

1. No, the reaction wheel can be smaller, but would have to be more massive to have the same moment of inertia. A hoop of denser material (steel, lead(!)) would be better.
The denser the better.

1. Rollyn01 says:

Depleted Uranium dense? That could be good for a very small reaction wheel and a large carriage load.

2. Montaray Jack says:

I’ve used Tungsten Carbide on occasion for just such a thing.
15.80 g/cm³ Tungsten Carbide
19.05 g/cm³ U-238 Depleted Uranium
19.25 g/cm³ W Tungsten
Just keep in mind its brittleness and the trouble it take to work with it. Diamond tooling and grinding. I ran Wire EDM tools and cutting carbide is not any tougher than cutting high speed steel, except for the carbides that use cobalt as a binder, EDMs can deplete the binding causing weakness near the EDMed areas.

I had the foundry sinter a form near the size and shape I needed. I asked the PhD Engineer who’s design this was for to check the numbers. We were already trying to fix a screwup in his design. Originally he called for steel for the counterbalance, it wasn’t dense enough and destroyed prototype #1. I couldn’t calculate the numbers myself since he was keeping that info secret. Plus he had the design in ProEngineer, I figured he had it right this time. What the heck, it’s only math from second semester Physics, and this guy is a PhD, what could go wrong?

I get the Carbide from the foundry, EDM some screwholes in the sinker EDM and then set it up in the Wire Machine. Ask the Engineer again to ok the numbers I was going to cut. I reminded that I could always cut something smaller, but I can’t cut it bigger. This was an expensive chunk of carbide. He gave me the go-ahead. 8 hours of the machine cutting or so later and we have a finished counterbalance. He leaves with his ne part.

A couple of days later, he’s back, counterbalance in hand. It wasn’t the right shape, or didn’t have enough mass far out enough and he tried to brase on another piece of something (not carbide I’ve no idea what it was) when it cooled the braise (with its much, much different coefficient of expansion) shrunk and shattered the finely machined piece of carbide into a million pieces. He wanted to know if I could fix it.

HA
HA ha ha

BTW this idiot won DesignNews’s Engineer of the Year Award for this “design”. What a joke.

6. Iw2 says:

Next step would be to implement steering by causing the “bicycle” to lean to one of its sides.

1. Nuno.S.Almeida says:

That’s awesome, but remember that this is a “guy in a garage”, not a multi-national with billion-dollar resources.

1. David K says:

Hello Almeida,
I saw a few comments you made were really great and useful. How do I find your posts or message you?

7. I love both videos. There are even more balancing robots on his profile…
:-)

8. rue_mohr says:

wickid! I want to make a full size version!

1. smilr says:

Did you by any chance miss the lines: “But why use two wheels when you only need one? His unicycle version can also be seen embedded after the break” in the article, or the second embedded youtube video?

1. Le Samourai says: