Does this project spark your interest?

Become a member to follow this project and don't miss any updates

Medicycle - Urban Responder

We are living in the future, Medics now need motorbikes to navigate our congested cities

15 43 32
Enjoy this project?
Share on twitter   Share on Facebook

This project was created on 05/14/2014 and last updated 3 days ago.

"Its the year 2350, flying cars never really took off and so the world is gridlocked. This is why we are introducing the Medicycle urban responder.

The Medicycle's use of revolutionary technology allows our emergency response teams to navigate traffic with ease whilst the latest up to date information is beamed directly to the riders heads up display from the central control centre."

The Medicycle is a self balancing twin tire unicycle, it will use an Arduino Mega in conjunction with an inertial measurement unit to keep itself balanced.
Useful information such as battery condition or even somebody's pulse can monitored with the heads up display which is wirelessly linked to the unicycle.

We have built self balancing machines before, the video can be found here


A) Self balancing electrically powered unicycle design, resembling a motorbike but with just one wheel.

B) Configured as a first response paramedic vehicle of the future so it will carry basic medical kit and monitoring equipment such as ECG, Blood pressure cuff, oxygen saturation monitor.

C) Crucially, these will be integrated into the electronics of the machine and the data on the patient will be sent via a mobile data network to the web, for example so that a remotely placed doctor or co-ordinator can i) offer advice and ii) this allows a receiving hospital to have real-time medical data on the patient before they even arrive.......a form of telemedicine.

D) Helmet or head mounted monocular display of vital parameters using a high contrast OLED display and prism arrangement. These can be performance data of the unicycle (e.g. Tilt angle, remaining battery life, torque commends being sent to the motor) OR data on the casualty being attended by the paramedic.

Aim is to demonstrate all these functions in an integrated machine by the end of the competition.

Each main module will be developed in parallel with build updates on each, they will be listed under the following headings.






Self Balancing control System for Unicycle 

Medical Pack and Monocular Display


a) The code for the self balancing on the Arduino Mega 1280

b) The code that runs on an Uno attached via serial link to the Mega, which then in turn runs the individual serial-enabled display modules on the front panel.

This lets the Mega spend all its time doing the self-balancing while the Uno can catch up running the displays in its own time.

Here is the location of this website.:

Project logs
  • Wireless Medical Data Transfer to the WEB

    4 days ago • 0 comments

  • Head Mounted Display Progress!

    4 days ago • 0 comments

    John and Dave are doing a great job, to get all the aspects of the project ready for the deadline is not easy task but we are almost there, this super cool head mounted display could have been a prize entry on its own!

  • Discovery Channel - Daily Planet

    22 days ago • 0 comments

    A couple of weeks ago a researcher from the Discovery channels Daily planet program stumbled across our 2min video for the HackaDay prize, they later contacted us asking if they could film the Medicycle in action.

    After altering a few holiday plans we met up in London and spent the day filming, the Medicycle perfomed flawlessly, the extra wide wheel even made it possible to ride across gravel without issue.

View all 16 project logs


Kevin wrote 16 days ago null point

Great project! I've been following you on Instructables for quite a while and have been impressed with your work.

If you don't mind me asking what does your potentiometer for overall gain actually adjust. I've been looking at the code and I only see one place where it is used. I know you indicated it adjusts tightness or sloppyness of the ride but could you expand a little more on what variables the knob effects and what this does to the output of the motor or anything it might control?

Keep up the good work!


Are you sure? [yes] / [no]

XenonJohn wrote 16 days ago null point

Hi Kevin,
The code is a bit messy I know but it does work. Think of the Overall Gain potentiometer like this: You set the P I and D constants via three potentiometers and from these and the gyro/accelerometer inputs you eventually get the amount of torque to be sent to the motor (updated 100 times per second).
The Overall Gain potentiometer simply takes this total value from the previous calculations and allows you to multiply it by 1, or maybe 1.2, or maybe a little more or a little less.
It is a crude way to adjust the feel of the machine just using one knob, but without altering the ratios of the P, I and D constants with respect to each other, the ratios of the P I and D to each other can remain unaltered therefore. Think of the P I and D constants as fine tuning of the balance characteristics of the machine, and the Overall Gain potentiometer as a kind of "volume" knob. It just makes the ride quality "tighter" or "mushier" by turning one knob, once you have (earlier on) managed to get the P I and D values right by a lot of trial and error.
Hope this helps, clearly it not particularly vital to have such a control knob at all.

Are you sure? [yes] / [no]

Kevin wrote 15 days ago null point

Thank you very much for the explanation!

I know you are very busy and your free time is limited.

Your efforts to share your knowledge and experience are appreciated.

Thanks again!

Are you sure? [yes] / [no]

voxadam wrote 2 months ago null point

How are the Medicycle and Nick's Raptor related? Is the Medicycle the evolutionary step? Do you expect to see an increase in top speed, carrying capacity, distance? Do you have target numbers for the Medicycle? About what does it weigh in its current incarnation?

In the current "dicycle" configuration is the Medicycle able to balance itself at rest without a rider?

One last thing (for now), why did you choose to use MY1020z motor instead of a hub motor? Hub motors are far more expensive, so, I feel reasonably comfortable assuming that was at least part of the decision. That being said, if all things were equal would you have gone with a hub motor?

Are you sure? [yes] / [no]

XenonJohn wrote a month ago null point

Many questions so many answers........

Nick's Raptor used an analog IMU and code I had written for my self balancing skateboard, modified for a one-wheeler. That sent motor speed commands via a serial link to a Sabertooth motor controller that generated its own pulse width modulated (PWM) power to the motor.
The mechanical design of the Raptor was extremely neat and all done by Nick, while still a teenager I might add.

At the same time I experimented with getting the self-balancing code working with a digital IMU as the analog ones are now harder to find, and also speeding up the (too slow) 500Hz PWM of the Arduino to a faster rate so I could then use it to drive an OSMC power controller which can handle huge currents. Having achieved these upgrades I tested them out in the Uni-Mig01 scooter which was a monowheel Vespa styled scooter with a really wide Kart tyre.

This control system is now in the MediCycle with further improvements.

This retro-styled scooter worked fine, the new machine is obviously styled like a motorcycle as it is much cooler way to go. Top speed is limited (for now) to reduce broken bones while I develop and debug it. Top speed could be much higher limited only by the courage of the rider and power/drive system. Carrying capacity is an interesting issue as again so long as motor has enough torque to correct a tilt, most of the time the weight acts vertically down through the wheel, so actually is a very efficient way to carry loads, like those women who choose to carry huge loads on their heads, rather than on their backs.

You can swing top part of frame left/right a few degrees as in the "enicycle" unicycle ( to move your centre of mass left/right of the wheel. This should help with steering and the Raptor did not have this. This feature is still being tested.

The twin tyred-wheel is functionally equivalent to the really wide tyre I had on the Uni-Mig01 scooter in terms of contact area with ground. Could be replaced with a wide car rim ultimately. You turn by leaning left/right and the inner tyre compresses slightly, this means its diameter is functionally less than the outer tyre, so machine will turn. As a means of steering it worked fine on the Uni-Mig whereas with a single narrow tyre the Raptor needed a lot more skill to steer.

Distance again is theoretically large, as when balanced and running on level ground these machines draw very little current; just to overcome rolling resistance of the tyre, drivetrain, and wind resistance i.e. less than an equivalent 2-or-more wheeler. The extra (large) reserve power (i.e. potential available torque at the wheel) is held in reserve for when you are falling over or going up a hill, when the extra power is required rapidly and effortlessly.

I have tried hub motors. Most nowadays are brushless. A cheap e-bike brushless power controller will not do as it has to be able to go as fast backwards as forwards, and switch between the two rapidly when balancing stationary. Suitable power controllers are available but expensive and indeed I have built a brushless motor Segway type machine with a very expensive Roboteq controller in the past

The main problem is that hub motors do not generate much torque at the wheel when stationary or at slow speeds (as they are optimised usually for bikes), which in a self-balancer is exactly when you need high torque the most. For a self-balancer it has to be able to generate high torque at the wheel effortlessly else it will not self-balance. Therefore I chose an OSMC (brushed) motor controller which at a reasonable price can handle 150Amps continuous, and a low cost but hefty Chinese motor that has a 1:6 reduction gear built in. This is required when using a motorbike wheel as without it you would need a huge chain sprocket on the wheel to get the correct gear ratio and torque (turning force) at the wheel.

I have learned about appropriate motor power requirements and gear ratio requirements for self-balancing simply from experience building these things. There is no "textbook" which is why this hobby is such fun. You learn something new with each machine you make.

It will balance itself at rest with no rider on it (with just a thumb on the deadman button of course)!

Are you sure? [yes] / [no]

PointyOintment wrote 2 months ago null point

How does it turn?

Are you sure? [yes] / [no]

XenonJohn wrote 17 days ago null point

We spent a whole day testing it recently on coarse gravel and also tarmac for a TV company. The steering pivot at the rear end that allows the upper frame to swing a few degrees to let you shift your weight to one side or the other over the wheel was locked solid for that day with a bolt and the handlebars were solidly welded to the frame. We found it can easily be steered on tarmac by just shifting your weight around so long as your turns are long and graceful, on gravel it took some learning how to do it but it was also possible. Tyre pressures make a big difference too (it would be really cool to one day make a machine steered by a computer continuously altering the tyre pressures on the fly, like some military vehicles can do to adjust to different terrain).
The steering linkage is now being worked on using the lessons learned from this testing day. Expect an update soon. As in previous posts, you learn the most about this subject by building and real life testing.

Are you sure? [yes] / [no]

rapfg4 wrote 2 months ago null point

Very impressive progress in such a short time!!

Are you sure? [yes] / [no]

XenonJohn wrote 2 months ago null point

A big update coming very soon. It runs and we have been developing software and redesigning hardware as a result of good old fashioned human experimentation! I have a Kevlar helmet which comes in handy.

Are you sure? [yes] / [no]

JackRC wrote 3 months ago null point

Just wondering how are you going to brake??
I mean if you hit the brake hard you will run over bar and hurt yourself really bad like I did a few months ago. I broke my neck and still recovering :(

Are you sure? [yes] / [no]

XenonJohn wrote 3 months ago null point

Hi Jack,
To slow down you lean backwards. The machine thinks it is falling over backwards so slows down, and if you persisted it would even go into reverse. All energy from this goes back into battery by regenerative braking. To go forwards you lean forwards, but alternatively there is also a twist grip handle on left handlebar then when twisted resets the target "balance point" so machine leans forwards and off you go. It acts as a twist-grip throttle in effect. I could develop something similar to make it slow down also.

Are you sure? [yes] / [no]

XenonJohn wrote 3 months ago null point

There are electric vehicle drag races so it would be cool to set up a new class for electric unicycles and have an inaugural event!
Am trying to build in a steering geometry similar to the one Aleksander Polutnik used in the enicycle. This should in theory make it much easier to learn to ride. We shall see.

Are you sure? [yes] / [no]

rapfg4 wrote 3 months ago null point

Yo, I am also currently working on a single wheeled, DC powered, self balancing motorcycle and we should set up a race when everything is complete. It would be a first ever!!

Are you sure? [yes] / [no]

loriffic wrote 3 months ago null point

Awesome project!

Are you sure? [yes] / [no]

Adam Fabio wrote 3 months ago null point

I'm a sucker for a self balancing vehicle! Thanks for entering mediacycle in The Hackaday Prize! Don't forget about the connected portion of the prize - though you've got plenty of technology already with your wireless heads up display!

Are you sure? [yes] / [no]