The SKATEway project is discontinued

The SKATEway project has been discontinued for administrative reasons. Its members will receive new assignments.

[SKATEway]Post 8


As says Alban, our project is currently suspended by th teachers.

We are waiting for the final decision. For the moment, I am redirected toward another branch.




SKATEway, week 7, 2/2


no further official news.


SKATEway, week 7 , 1/2


Charles and I have been working quite intensely on our exams, so we did not make any progress.

However, our project is not going well at all, therefore Sam & Alexis quite understandably do not want to have anything to do with it any more. As things are going, we are driving straight into a wall. We thus must reflect on what we want to do now in order to be able to take a fresh start and actually achieve something this year. More about this in the following days.



[SKATEway]Post 8


Unfortunately the end of the week was as productive as the begin. I have not enough time to keep the project above the surface. I am currently in a compressed work week with three exams. I know it’s not an excuse. So this week in addition to the challenge, I only order this IMU 6 Degrees of Freedom to test the angle capture.


[SKATEway]Post 6


This 3 days I essentially worked on the diagram bloc to self balance our board with a null angle to the horizontal.

Its a simple PID but we have to add some stuff to make the regulation working. First we need to prevent the wind up on the integrator when the output is satured. Then the D componet increase hight frequencies noise. So we attach a low pass filter. FinaLly we have the value to apply on the motor. The board move to an other angle. Perturbations can emerge. Like the ground slope, the wind or humans. We got the final position now. We need accelerometer and gyroscope output to calculate the new angle of the board to the horizontal. We can directly use the output of the accelerometer but we have to apply an anti drift filter to the gyroscope output to cancel the integrated error over the time. Then we compare the request angle with this angle and do the operation again.

Here, the diagram bloc !



SKATEway, week 4, 2/2


We’ve had quite a tough week, Charles and I. Because we had some trouble working together, our teachers had to shake us up to get us to work efficiently. I believe this was necessary, and it allowed us to set a better working dynamic.

I studied the motor with the help of Alexis, and we now know exactly what we are looking for: a geared brushless hub engine with a custom built controller. To build the controller, we will use VESC . We will have to build it on a seperate PCB for two reasons: protect the main PCB from this one and vice versa, and get better task splitting during the project, which will increase our efficiency. I am in charge of studying the motor and VESC part. It has to be done by tomorrow evening.

We also know that we need an IMU and 2 angular sensors to get each foot-pad’s angle from the main bloc, but we still don’t know how to install those sensors on the board. I am also in charge of this.

This week has been absolutely exhausting because of the stress of the meeting with our teachers on friday evening, but it allowed us to kick it into the next gear, so hopefully everything will be fine now.


[SKATEway]Post 4


This week we got a scolding for our teachers because we have not worked a lot so far. We have to stand up and move on.

Now we have to concretize the choice of sensors and study about the regulation of the system.

On my side I will study about regulation. We already know we will use a closed loop because the output feedback on the system. Learn what is already existing. Chose several algotithm PID, SS, SIMO, MIMO, RST, LQR. Then determine the number of instruction to estimate the worst execution time of the loop, and the system’s response.

I know PID is the simple but not the more efficient, to be seen.

Determine what is the input of the regulation, an angle (not necessarly null), a speed. Think wich angle will we use. Absolute horizontal or about the road.




[SKATEway]Post 3

Hi, from now we have to post Wednesday and Sunday for more follow up.

We finally have all our component except the micro-controller. Motors are not final too. Because it is hub. It is compact but not necessarily enough powerful.

For the moment we have :

We are waiting for the teachers agreement to begin our PCB.




SKATEway, week 3


This week was a rather strenuous one. Indeed, what we thought to be a reasonably difficult problem has turned out to be a bit too much for us. Indeed, calculating the necessary torque for our skateway should be possible, especially since it is quite like the classical exercise of the dynamical study of an inverted pendulum. And yet, the equations seem to always elude our grasp.

Newton’s 2nd Law yields this system of equations:

\left\{\begin{matrix}  (m + M)\ddot{x}-ml\ddot{\alpha }cos(\alpha )+ml\dot{\alpha }^{2}sin(\alpha) = \frac{C}{r} \\  l\ddot{\alpha }-gsin(\alpha ) = \ddot{x}cos(\alpha )  \end{matrix}\right.

In which:

  • m : mass of the user
  • M : mass of the SKATEway
  • l : height of the center of gravity of the user
  • alpha : angle at which the user is tilted (0 if standing straight)
  • g : gravitational acceleration
  • r : radius of the wheels
  • C : torque of the engine

However, how do we extract the necessary torque from this? What exactly is our scenario? Is it: going from the maximum value of alpha to standing straight? Is the skateway moving or not? Are we simply trying to keep the user in the unstable equilibrium in which the acceleration of the skateway is compensating gravity?

Many questions with few answers.

A very crude approximation is to consider the device as a lever instead of an inverted pendulum: it is equivalent to blocking the wheels. In this scenario, the torque is independent from the radius of the wheels, and is given by the formula: C = m(l+r)gsin(\alpha). However crude this may look, it is the best we could do.

I feel like I am missing something obvious. It should not be that hard, it is a very common problem, the tools to use are well-known, and yet, each and every time, the answer slips away from my mind. Any help on this would be immensely appreciated.

Where does that leave us? One week behind schedule. We did find DIY LiFePo batteries as well as what appears to be a satisfying hub engine, although they were not vetted by our teachers yet. Because we do not have a torque value precisely calculated, we are not too sure of our estimation. Since we need to move on, we will probably use this crude approximation; we did compare it to existing projects, such as gyropodes, and it is of the same order of magnitude. And now, onward with the vetting and then, the liaison.

Edit: with the crude approximation of C = m(l+r)gsin(\alpha), we get: C ~= 311 N.m.

If we assume a maximum speed of 25 km/h (the same as the oneWheel), we need 600 RMP.

Such an engine would require about 20kW. This is a riduculously high value which we cannot use.