Skip navigation
The science of balancing a bike

When a human rides a bike, he or she uses almost all the sensory information available to control the vehicle, starting out with what we call proprioceptive information; the rich sensors that are in our arms and limbs.  For example, I know what my arm is doing now but I can't see it and it turns out that's an important feedback variable for humans to control dynamic systems.

Right now there are lots of companies that design bicycles but there is no scientific understanding of how the parameters of the vehicle affect the dynamics and affect the ability of the human to control.

So, this is one of the two bicycles that we've built for the research project. And this one is set up so a person actually rides it and can control the bicycle in a restricted way so that we can understand the dynamics, the motion and what the riders are doing.  

I would say that the project got started for me in my senior year of undergraduate.  I was on a team where we were tasked with building a recumbent bicycle and if you've ever ridden  recumbent bicycle, you know that it's more difficult to ride than a normal bicycle, and so we were trying to figure out, well where do we put the mass?  How do we situate the person on there?  How do we adjust the geometry of the bicycle?  And I realized at that point that there were no guidelines to allow you to do that.  Certainly not in the engineering way that we're trained to do.  

So that was sort of the genesis of my curiosity about what's going on with the bicycle and why would one bicycle be harder to control than another and how can we tune the vehicle to make it easier to balance?  

So we've got two main experiments that we're doing.  We're doing them both in the gymnasium and also on a treadmill and what they are, are one is a lane-change maneuver and the other is what we call like a disturbance input. In the lane-change maneuver, I'm looking at the line on the ground and I'm trying to keep my front wheel on that line as we go.  But in that case I only have the control input, is only what I do with my arms and the steer.  The other set up is where we also try to track a line on the ground.  It's just a straight line and then we have an external disturbance.  It would be very similar to if a wind gust blew you or somebody came up and just pushed you.  In our case we're pulling the bicycle from an external disturbance and in that point we're measuring two inputs.  We're measuring what the rider does to control and then also the sort of external input from the person riding beside the bicycle.  

There's no scientific understanding of why people do feel comfortable when they do feel comfortable on a bicycle and why they don't when they don't.  So, that's one of the things that we're trying to understand is. How you would design a bicycle for a certain person, perhaps or for more likely for a certain population of individuals like old folks or kids or young adult athletes and how they should be different to make those different populations feel comfortable.