BNO055 Y-axis linear acceleration drifts after a while

[Justin]

Hello,

We have been using bno055 sensors to get data about footsteps, so we have been testing it and acquiring data in NDOF mode and got the different linear accelerations values out of it. Once, I was walking for a long time, and all seemed fine except at one moment the Y-axis linear acceleration started drifting. Then, after a break where I didn’t walk it started working fine again.

I’ll join the graph that was made out of this test, so you understand what I am talking about.

Does anyone have any idea why it is doing that?

Thank you in advance.

[BSTRobin]

Hello Justin,

Could we know what was the parameters of the sensor you configured?

[Justin]

Hi Robin, thank you for answering.

I hadn't changed the paramaters from the default configuration of the BNO055 when I made this test, so the paramaters should have been those according to the datasheet :

Does it answer your question ?

[BSTRobin]

Hello Justin,

Ok, got your parameter configuration.

Is it easy to reproduce it if we test it?

[Justin]

Hello ,

Basically, I've wrapped the sensor around my ankle and walked outside with it  for periods of 5 minutes or so and stopped walking in between. 
Then I've extracted the data and plotted the linear accelerations values.

So yeah, it is simple to reproduce.