MPU-9250 (MPU-6500 + AK8963) I2C Driver in Python
MPU-9250 is a multi-chip module (MCM) consisting of two dies integrated into a single QFN package. One die the MPU-6500 houses the 3-Axis gyroscope, the 3-Axis accelerometer and temperature sensor. The other die houses the AK8963 3-Axis magnetometer. Hence, the MPU-9250 is a 9-axis MotionTracking device that combines a 3-axis gyroscope, 3-axis accelerometer, 3-axis magnetometer and a Digital Motion Processorâ„¢ (DMP). The hardware documentation for MPU-9250 can be found at Product Specification and Register Map and Descriptions.
Table of Contents
- Instalation
- How To Use
- Getting Data
- Calibrating Sensors
- Reset Registers
- Final Notes
- Sponsors and Donations
- What We’re Working on Now
- How To Cite
- License
Instalation
To install via pip use:
pip install mpu9250-jmdev
To use the package for development purposes use:
git clone https://github.com/Intelligent-Vehicle-Perception/MPU-9250-Sensors-Data-Collect.git MPU9250
cd MPU9250
pip install -e .
pip install -r requirements.txt
Known Issues
:information_source: Notice that this package requires Python 3.6 or higher.
:warning: If you have both Python 2 and 3 installed on your machine, use
pip3
to install andpython3
to run instead.
:warning: If you run your python source-code with
sudo
, remember to usesudo
withpip install
commands as well.
:warning: If you get zero values and the message Using Fake SMBus, the smbus2 requirement has not been installed correctly. Try installing manually with the command
pip install -r requirements.txt
orpython -mpip install smbus2
.
:exclamation: Any other problem or questions, open an issue in this repository clicking here, do not send me an e-mail!
How To Use
With I2C Bus, you can use the MPU-9250 in two ways: simple mode or advanced mode. The example source-codes are in examples folder.
Simple Mode - Master Only
In this mode, the MPU-9250 connects directly to Raspberry GPIOs. There are two physical addresses available for the MPU-9250, being 0x68 and 0x69. Therefore, on each I2C Bus you can have up to two MPU-9250 connected. The connection between GPIOs and MPU-9250 is as follows:
MPU9250 | Raspberry | Note |
---|---|---|
VDD | 3.3V | On some models of the MPU-9250 5V can be used. |
AD0 | 3.3V | If used, the MPU-9250’s address is changed to 0x69. Otherwise, the address is 0x68. |
GND | GND | Â |
SDA | SDA | Â |
SCL | SCL | Â |
Below simple code to test the execution with never ending loop:
import time
from mpu9250_jmdev.registers import *
from mpu9250_jmdev.mpu_9250 import MPU9250
mpu = MPU9250(
address_ak=AK8963_ADDRESS,
address_mpu_master=MPU9050_ADDRESS_68, # In 0x68 Address
address_mpu_slave=None,
bus=1,
gfs=GFS_1000,
afs=AFS_8G,
mfs=AK8963_BIT_16,
mode=AK8963_MODE_C100HZ)
mpu.configure() # Apply the settings to the registers.
while True:
print("|.....MPU9250 in 0x68 Address.....|")
print("Accelerometer", mpu.readAccelerometerMaster())
print("Gyroscope", mpu.readGyroscopeMaster())
print("Magnetometer", mpu.readMagnetometerMaster())
print("Temperature", mpu.readTemperatureMaster())
print("\n")
time.sleep(1)
Advanced Mode - Master-Slave
If you want to have more than two MPU-9250 on one I2C Bus, you must use Master-Slave mode. In this case, first configure the MPU-9250 according to the previous section, they will be used as Master. To configure the MPU-9250 Slaves, connect as follows:
MPU9250 Slave | MPU9250 Master | Raspberry PI | Note |
---|---|---|---|
VDD | Â | 3.3V | On some models of the MPU-9250 5V can be used. |
AD0 |  | 3.3V | If used, the MPU-9250’s address is changed to 0x69. Otherwise, the address is 0x68. |
GND | Â | GND | Â |
SDA | EDA | Â | Â |
SCL | ECL | Â | Â |
This way you will have an MPU-9250 Master connecting SDA and SLC directly to the GPIO in Raspberry PI, and an MPU-9250 Slave connecting SDA and SLC to the EDA and ELC in MPU-9250 Master.
Below simple code to test the execution with never ending loop:
import time
from mpu9250_jmdev.registers import *
from mpu9250_jmdev.mpu_9250 import MPU9250
mpu = MPU9250(
address_ak=AK8963_ADDRESS,
address_mpu_master=MPU9050_ADDRESS_68, # Master has 0x68 Address
address_mpu_slave=MPU9050_ADDRESS_68, # Slave has 0x68 Address
bus=1,
gfs=GFS_1000,
afs=AFS_8G,
mfs=AK8963_BIT_16,
mode=AK8963_MODE_C100HZ)
mpu.configure() # Apply the settings to the registers.
while True:
print("|.....MPU9250 in 0x68 I2C Bus - Master.....|")
print("Accelerometer", mpu.readAccelerometerMaster())
print("Gyroscope", mpu.readGyroscopeMaster())
print("Magnetometer", mpu.readMagnetometerMaster())
print("Temperature", mpu.readTemperatureMaster())
print("\n")
print("|.....MPU9250 in 0x68 I2C Bus - Slave in 0x68 auxiliary sensor address.....|")
print("Accelerometer", mpu.readAccelerometerSlave())
print("Gyroscope", mpu.readGyroscopeSlave())
print("Temperature", mpu.readTemperatureSlave())
print("\n")
time.sleep(1)
Getting Data
All sensors and measurement units of the MPU-9250 are described below:
Sensor | Unit |
---|---|
Accelerometer | g (1g = 9.80665 m/s²) |
Gyroscope | degrees per second (°/s) |
Magnetometer | microtesla (μT) |
Temperature | celsius degrees (°C) |
Before read the sensor data, make sure that you have executed the command:
mpu.configure() # Apply the settings to the registers.
Reading Accelerometer
The accelerometer measures acceleration in three axes (X, Y, Z). To read your data, use the commands:
masterData = mpu.readAccelerometerMaster()
slaveData = mpu.readAccelerometerSlave() # If there is a slave
Reading Gyroscope
The gyroscope measures rotation rate in three axes (X, Y, Z). To read your data, use the commands:
masterData = mpu.readGyroscopeMaster()
slaveData = mpu.readGyroscopeSlave() # If there is a slave
Reading Magnetometer
The magnetometer measures geomagnetic field in three axes (X, Y, Z). To read your data, use the command:
masterData = mpu.readMagnetometerMaster()
When used in Simple Mode (Master Only), the magnetometer will be available on the I2C Bus with address 0x0C. When in Advanced Mode (Master-Slave), the magnetometer will also behave as a slave, and address 0x0C will not appear on the I2C Bus, acting as an auxiliary sensor.
Reading Temperature
The temperature sensor measures data in Celsius degrees. To read your data, use the command:
masterData = mpu.readTemperatureMaster()
slaveData = mpu.readTemperatureSlave() # If there is a slave
Reading All Data
If you want to read data from all sensors (master and slave) at the same time, use the commands below (useful for saving to csv):
labels = mpu.getAllDataLabels() # return labels with data description for each array position
data = mpu.getAllData() # returns a array with data from all sensors
Reading All Settings
If you want to read settings (biases, resolutions, scale factors) from all sensors (master and slave) at the same time, use the commands below (useful for saving to csv):
labels = mpu.getAllSettingsLabels() # return labels with settings description for each array position
data = mpu.getAllSettings() # returns a array with settings from all sensors
Calibrating Sensors
This library has functions ready for calibration accelerometer, gyroscope and magnetometer sensors. To calibrate all sensors at once, use the command:
mpu.calibrate() # Calibrate sensors
mpu.configure() # The calibration function resets the sensors, so you need to reconfigure them
Accelerometer and Gyroscope
To calibrate the accelerometer and gyroscope sensors, make sure that the sensors remain fixed and stationary. Align the accelerometer’s Z axis with gravity, i.e., gravity (1g) should only appear on the sensor’s Z axis (place the sensor in a flat place). To perform calibration run the command:
mpu.calibrateMPU6500() # Calibrate sensors
mpu.configure() # The calibration function resets the sensors, so you need to reconfigure them
abias = mpu.abias # Get the master accelerometer biases
abias_slave = mpu.abias_slave # Get the slave accelerometer biases
gbias = mpu.gbias # Get the master gyroscope biases
gbias_slave = mpu.gbias_slave # Get the slave gyroscope biases
The biases are programmatically applied to the sensor data. Therefore, when reading the sensor data, the biases will be applied internally, returning corrected data. If you have calculated the biases of these sensors once, and want the controller to use them, simply parameterize as follows:
mpu.abias = [0, 0, 0] # Set the master accelerometer biases
mpu.abias_slave = [0, 0, 0] # Set the slave accelerometer biases
mpu.gbias = [0, 0, 0] # Set the master gyroscope biases
mpu.gbias_slave = [0, 0, 0] # Set the slave gyroscope biases
Magnetometer
To perform calibration run the command:
mpu.calibrateAK8963() # Calibrate sensors
mpu.configure() # The calibration function resets the sensors, so you need to reconfigure them
magScale = mpu.magScale # Get magnetometer soft iron distortion
mbias = mpu.mbias # Get magnetometer hard iron distortion
If you have calculated the biases of these sensor once, and want the controller to use them, simply parameterize as follows:
mpu.magScale = [0, 0, 0] # Set magnetometer soft iron distortion
mpu.mbias = [0, 0, 0] # Set magnetometer hard iron distortion
Reset Registers
If you want to reset the values in all registers of all sensors in all MPU-9250, execute the command below:
mpu.reset() # Reset sensors
mpu.configure() # After resetting you need to reconfigure the sensors
Final Notes
The folder mpu9250 consist of the high level library. The folder examples contains files with basic execution and threaded examples.
Sponsors and Donations
This project does not have any funding. To help maintain the project, consider making a donation 🙌.
What We’re Working on Now
New features are currently under development:
- V2: new major version will be released soon, much more robust than the current one. It will supports several MPU models, such as 9250, 9150, 6050, 6500.
- Complete support documentation for configuring RPi, I2C, VNC, SSH, etc.
How To Cite
To cite this repository, use the reference below:
@software{menegazzo3960441,
author = {Jeferson Menegazzo and Aldo von Wangenheim},
title = ,
month = jul,
year = 2020,
publisher = {Zenodo},
version = {1.0.12},
doi = {10.5281/zenodo.3960441},
url = {https://github.com/Intelligent-Vehicle-Perception/MPU-9250-Sensors-Data-Collect/}
}
License
This project is under Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). Please see License File for more information.