Bosch Sensortec Community

    Showing results for 
    Search instead for 
    Did you mean: 

    BNO055 vs BHI260AB fusion accuracy

    BNO055 vs BHI260AB fusion accuracy

    Long-established Member

    Hi all,

    Some context of the project I'm working on

    To give some context to my question, let me give some information about the project we are developing.

    We are developing a device that will be used in the medical/surgery market and our main need is to have an IMU that gives us relative rotation (we have a calibration routine for setting the reference measurement) information in quaternions.

    After looking at the Bosch portfolio, we realized that the best option would be the BNO055, but because it's going out of stock in most of the suppliers we are considering using the BHI260AB.

    The question

    As far as I know, BNO055 runs internally the BSX3 full library, while the BHI260AB the BSX4 full library.

    I'd like to know if there's a difference in the final output accuracy of the sensor fusion between these two chips used in IMU mode.

    The sensor fusion accuracy is specified for the BHI260AB, but I couldn't find the same information for the BNO055. In the case of the BHI260AB:

    BHI260AB Sensor fusion performance:

    – Static accuracy (head., pitch, roll): 2, 2, 2 degrees

    – Dynamic accuracy (head., pitch, roll): 7, 2, 2 degrees


    As I mentioned before, I've seen in a post that there's no difference in the accuracy of the sensor fusion between BSX3 full and BSX4 full. But, since BNO055 and BHI260AB have some differences in the characteristics of the sensors, that may imply that the final fusion output accuracy is also different.


    4 REPLIES 4

    Community Moderator
    Community Moderator


    Thanks for your inquiry.

    It is true that BNO055 and BHI260AB have the same sensor fusion accuracy for heading/pitch/roll and quaternions. The main differences between them are as below.

    1. BNO055 is fixed, while BHI260AB is programmable by using the SDK.

    2. BNO055 consumes much high current consumption than BHI260AB.

    3. BNO055 sensor fusion output data rate (ODR) is up to 100Hz, while BHI260AB up to 800Hz. So you may apply your own filtering algorithm on BHI260AB at 800Hz ODR to get better accuracy of Euler angles and quaternions.


    Long-established Member

    Thank you very much for your prompt reply!

    About the BHI260AB fusion accuracy specified (which is the same as the BNO055), are those absolute errors? and in which conditions were those errors measured?

    Thank you!


    Yes, those are absolute errors for Euler angles H/P/R. We measured these absolute errors by operating the test device in a wooden room where there is no magnetic interference. In the room there are total 12 cameras installed at certain locations around the roof facing the test device. The wooden room has known magnetic north direction so that the camera system is calibrated to north. Whenever the test device is rotating or stationary, the camera system can determine the dynamic and static angles of the test device. The errors will then be determined as the subtraction between camera system and BHI260AB or BNO055 Euler angle outputs.


    Long-established Member

    Excellent! Thanks!