Single wheel robots

Wheels… Most of our inventions that have to move use them. Just think about it! Since the invention of wheel we have made numerous types of carts and carriages, our cars have wheels, our vacuums have wheels, our bicycles have wheels.

The reason behind this is quite obvious – the wheel is often the most cost-effective and effective solution to different movement problems. Put at least three wheels in a polygonal formation and voila – you have a statically stable platform able to move.

Single-wheel platforms, on the other hand, are not statically stable – they fall without support. To provide stability and usability you’d have to solve quite challenging engineering problems. So, why bother?

Firstly, solving complicated engineering problems is good for humanity. Secondly, dynamically stable single wheel robots have a potential to become very agile and compact. Thus – ideal to use in crowded environments in interaction with humans.

Of course, usually we imagine humanoid robots in this role. But let’s face it – often it would not be necessary to use a humanoid if its unique abilities are not explicitly required. A single wheel robot could be a cheaper solution for a robot receptionist who has to work on one floor only.

Murata girl

Murata is around 60 years old Japanese company who operates in electronics field. It manufactures various electronics components. Murata robot prototype line is something like a showcase of what can be done using company’s products.

Murata girl, also known as Murata Seiko-chan, is a robot that basically rides a unicycle. According to, the robot has gyro sensors to provide stability measurements, an ultrasonic sensor for object detection and a bluetooth module for wireless communication.

Quite cute, isn’t it? Of course, this is not designed with a specific application in mind. However, maybe someday Murata engineers will use knowledge obtained during this project to make something really applicable.


At the beginning of this article I mentioned that single wheel robots could be useful for applications where interaction with humans is required. Ballbot is a prototype built to demonstrate this ability. It is tall, quite thin and it balances on a ball.

To build a robot of this height using other, more conventional methods, you’d have a few options. You can build a statically stable multiple-wheel robot. To do this you’d have to use a large base with a low center of gravity. These robots are not very agile, as they can tip over when accelerate or when they are on a slope. This could make them a bit cumbersome.

Another option is to make a humanoid. This option also has its flaws. Humanoids have many joints, thus they require many motors, thus they require much power, thus they require powerful batteries that are either heavy or expensive or both.

Apart from these issues, it seems more likely that a robot such as Ballbot will be agile enough to become widely used. So, it has an onboard computer, a measuring unit with gyro sensors to measure angles and a drive unit.

A drive unit is something like those now rarely used computer mice (or mouses, or mouse devices, whichever you prefer) with a ball. You move the mouse, thus moving a ball, which moves rollers, which encodes information about the direction and speed of movement. Don’t say you haven’t disassembled one. Well, this robot works in a similar manner. Only this time – rollers move the ball, thus moving the robot.

I can almost hear you shouting – “What about stairs?” Well, stairs are an issue for this and other single wheel robots. However, most office buildings, hospitals, malls and other places where they could operate have elevators and moving staircases. This kind of robots could successfully use those.

Initially, the idea was embodied at Carnegie Mellon University (What a surprise, eh?). Now, students and researchers at numerous universities around the globe have made their own ballbots (the name stuck). Some of these are – Tohoku Gakuin University and The University of Adelaide. I guess this is how progress works.


I kind of doubted if I should write about this prototype here. This article was supposed to be about robots that have a single wheel. This robot instead is a single wheel. I also plan to write a page on mobile spherical robots – robotic balls if you will. As this “wheel-robot” is relative to them, I could write about it there. Still, at the end I decided that this place is more appropriate for it.

So, the ingenious idea behind Gyrover is an observation that wheels express dynamic stability when in movement as opposed to multiple-wheel platforms. Thus, such robot could be more practical in some occasions.

For example the second prototype – Gyrover II was able to float and it was controllable on the water. So, this design could be used as an original amphibious solution. Also, robot’s vulnerable parts are enclosed in this “wheel”, so it can successfully resist exposure to dust, hazardous materials or environments.

The Gyrover was initially designed in Carnegie Mellon. It is gyroscopically stabilized so it can stay upright when it is not moving. Gyrover has a drive motor with an internal pendulum that enables forward movement, a motor that drives the gyro and a motor that tilts the gyro to enable steering.

Look at this video, to get a feeling of what it’s like:

You can find more info about Gyrover and the ongoing research at the project’s website. If you’re interested in the research side of this project I strongly encourage you to read the publications there.

As you see, single wheel robots are nowhere near actual applications as far as I’m aware. However, these concepts are really interesting and will bring fruits in foreseeable future. Maybe the moment when you’ll routinely chat with a ballbot-receptionist is not as far away as it may seem at first.

Go back from “Single wheel robots” to “Types of robots”.

Go back to home.