Using two metamotion R simultaneously, huge difference in timestamp
Hi,
In my bio-mechanics application I require 3 sensors for monitoring the activity.
For calculation of joint angle, it requires Roll (IMU1)- Roll (IMU2). Since there is huge difference in timestamp of two sensors , I am not getting accurate angles.
Can I reduce sensor fusion rate to 50Hz (even that is acceptable ) but the timestamp should be at regular intervals.
I have attached data from two sensors. (Epoch Time for first reading has been taken 0)
Comments
Please find the attached data. Also there is sudden jump in yaw reading, even after caliberation.
I have used pandas lib as suggested in the tutorials to synchronize the datasets of two sensors but it results in error in calculated joint angle.
Kindly suggest.
Awaiting your reply.
Thank you
Simmi
I believe that this has something to do with euler singularities but i am not sure.
Timestamped are synched to the bluetooth device used to download the data from the sensor. If you use two different devices and they have different clocks, you could see disparities between final timestamps.
Also, try to get more data samples and see if this is repeatable or just a one-time-thing.
Also, fix your sync + joint calculation code first; this also could be an issue.
Thanks Laura.
In my application, I need samples at constant rate and same time stamp. I will send you more datasets where the sensor is kept stand still . Yet there is error, in timestamps.
Kindly suggest the method for synchronisation of two sensors.
Again the timestamp is created by the device used to download the data, for example, if you use an iPhone and our MetaBase App, the timestamp is created by the iPhone clock. Please think about that when you are working on your App; the sensor has nothing to do with it. Placing it flat and recording will accomplish nothing as this is an issue that is smartphone/gateway hub side/bluetooth side.