Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Absolut lean angle instead of degree/second

edited March 9 in Feature requests
Hi aol,
I was looking into logging device lean angle for a motorcycle and apparently racechrono reports on lean angle speed (degree/second). Would it be possible to report on absolute lean angle in degree instead? This would be much more useful for motorbike track analysis.
Thanks

Comments

  • That's on my TO-DO list. There is theoretical calculated lean angle in GPS channels, but I guess you're after real value.
  • Yes and I've played around with the device lean angle, it seams pretty accurate. I want to fix the phone horizontally on the bike and leverage that if possible. Any ETA? Is it far down the list?
  • aolaol
    edited March 9
    My plan is to write my own algorithm for that, and not to use what the device is reporting. I'll start testing my own algorithms during when season begins, but I think it needs quite a bit of work. So the ETA is 'depends'... But it is high on the list.
  • I understand. When you say own algorithm, do you mean getting the raw data from device gyroscope and analyse it yourself? Isn't that data outputted in a standard format?
  • Gyroscopes output in degrees/second and you have to do a sensor merge with magnetometer. Devices do this themselves too but that algorithm is not meant for vehicle physics, so they will give false readings on certain situations.
  • Ok good to know.
    I'll monitor the chanlogs, thanks for the feedback
  • My microcontroller based GPS + logger I use Sebastian Madgwick's Gyro + accelerometer + magnetometer algorithm and returning pitch, yaw, roll in RC3$ values for logging in RC. I won't be testing until riding season starts as well.

    My accelerometer has a built in IMU system but I had issues utilizing it and found it was actually faster to get raw values and have the microcontroller do the calcs.

    @aol: why do devices not return the correct readings on vehicles? Wouldn't the algorithms all basically be the same?

    Jeff
  • aolaol
    edited March 26
    @J_D_W mainly because on motorcycles you cannot use gravity to compensate gyro drift, as gravity always points down. This is not saying that I would be expert on the matter, so I'll be happy to research any papers / algorithms / etc you have for me. I'll be looking at the one you named for sure!
  • @aol I see what you mean. Had not thought about that. But would it not be similar to quadcopters though as they 'lean' in corners.

    Now I want to get the sensors on the bike even more and see what I get

    Jeff
  • @J_D_W, I haven't looked at Sebastian M.'s algorithm yet, but the one that I tried first was from quadcopters and it indeed used an accelerometer to compensate gyro drift. But then again it didn't have magnetometer, so there was no other choice.
  • Hi aol,
    I've collected GPS lean angle last friday on a track day and it looked fairly accurate. I can send you a youtube link if you want to check it out. Any chance you could add a nice visualisation for overlay instead of the numerical value? maybe something in that spirit: https://i.ytimg.com/vi/i1qxHsAZsQI/hqdefault.jpg
    Thanks
  • @hahaha, yep this is coming soon
Sign In or Register to comment.