SmartMotion® MEMS Motion Sensors

Hello,
I am working on DK-20648. I am using 3 gpio pins i.e. pin 29, 30, 31 and I have set pin levels to 0. When there is interrupt these pins has high value. But when I ON the circuit, these pins are initially high and after some time, they have LOW value. So what is the solution for this? How these gpio pins has LOW value at power on also?

Thanks

Hello,

I'd like to know more about the status of the current implementation of the SmartMotion eMD library fo the motion sensors icm-20789/icm-20689

The current implementation downloaded [1] (20789 (DMP) SmartMotion eMD 1.1.0) is fairly incomplete specially regarding the DMP features. We can see which features are currently available from the inv_icm20789_sensor definition (in Icm20798Ctrl.h)

Hello,

I'm working with the icm20948 and the icm20689 and I'm using the EMD lib (v3.8.9).
I'm wondering what is the exact definition of the following sensors type (defined in SensorTypes.h and Icm20948Setup.h):
* INV_ICM20948_SENSOR_GYROSCOPE_UNCALIBRATED
* INV_ICM20948_SENSOR_RAW_GYROSCOPE
* INV_ICM20948_SENSOR_GYROSCOPE

I supposed 'raw gyroscope' stands for raw gyroscope data without offset correction. But what about the others 'uncalibrated' and 'gyroscope'?

Best,

Hi,

In an earlier post I was trying to understand the capabilities of the DMP, particularly around setting a float-based mounting matrix. Reading through the source code and the seemingly non-existent, non-sensical documentation wasn't getting me anywhere, so as a last attempt to make use of the ICM-20948 I bought the DK-20948 evaluation kit.

What are the units of Rotation vector (INV_ICM20948_SENSOR_ROTATION_VECTOR) accuracy (inv_icm20948_get_rv_accuracy()) on ICM-20948? Android is expecting this value to be in radians, but what I read it is approx. from 0.035 to 0.06 (in degree 2° to 3.43°) which is too low to be true. I need to interpret it as compass accuracy.

Hi all,

My project is using the ICM-20648 chip and I set it up for DMP output (32-bit 6-axis quaternion).

The IMU chip is installed on the device that is parallel to a flat surface.

It appears if I don't power up the device sitting parallel to a flat surface, the calculated yaw, pitch and roll will always be wrong. It will use the angled position as the "zero" reference.

Is there something I missed or is that a DMP requirement to have the IMU startup on a flat surface?

Hi,

We are using ICG-20660 device in our application and trying to read the GYRO data but we are getting data around 16'hFFFF and varies between low values like 16'h0020 and 16'hffbf.

We are not sure whether the data is signed or unsigned or two's complement as it is not mentioned in the datasheet, unlike accelerometer data which is in Two's Complement.

We have configured the following register

0X6B --> 0X01
0X1B --> 0X10

Hi,

In the datasheet of MPU6000 it's very well explained how to read data in burst mode using I2C.

However, such documentation is lacking for the SPI case. It's simply mentioned that "burst read" is supported, but there's no explanation on how to read the data. On I2C it's stated that the MPU6000 will automatically increment the register address to read data for more registers, does the same apply to SPI? What data should the MCU send to the MPU6000 in order to read in bursts (similar to the ACK in I2C)?

Thanks!

Does TDK offer software that would run on a generic ARM Cortex M4 for the ICM-20948 that provides Soft Iron Compensation?

Hi,

I'm using an ICM-20948 in my design, and I plan on using the DMP for sensor fusion computation. I've poured over all available technical documentation (datasheets, EMD user guides, etc.), but still find myself somewhat confused as to the various DMP functions. I've got the DK-20948 Atmel Studio project open, which allows me to jump between the various function calls, but I'm still unclear as to how the mounting matrix is established.

In sensor.c of the DK-20948 Atmel project, the below line (with comment) seems to point to what I'm trying to achieve: