mpu6050

no_std driver for the MPU6050 6-axis IMU

Datasheet | Register Map Sheet

Visualization options for testing and development in viz branch

Basic usage

linux_embedded_hal example

use mpu6050::*;
use linux_embedded_hal::{I2cdev, Delay};
use i2cdev::linux::LinuxI2CError;

fn main() -> Result<(), Mpu6050Error<LinuxI2CError>> {
    let i2c = I2cdev::new("/dev/i2c-1")
        .map_err(Mpu6050Error::I2c)?;

    let delay = Delay;

    let mut mpu = Mpu6050::new(i2c, delay);
    mpu.init()?;
    mpu.soft_calib(Steps(100))?;
    mpu.calc_variance(Steps(50))?;

    println!("Calibrated with bias: {:?}", mpu.get_bias().unwrap());
    println!("Calculated variance: {:?}", mpu.get_variance().unwrap());

    loop {
        // get roll and pitch estimate
        let acc = mpu.get_acc_angles()?;
        println!("r/p: {}", acc);
        
        // get roll and pitch estimate - averaged accross n readings (steps)
        let acc = mpu.get_acc_angles_avg(Steps(5))?;
        println!("r/p avg: {}", acc);

        // get temp
        let temp = mpu.get_temp()?;
        println!("temp: {}c", temp);

        // get temp - averages across n readings (steps)
        let temp = mpu.get_temp_avg(Steps(5))?;
        println!("temp avg: {}c", temp);

        // get gyro data, scaled with sensitivity 
        let gyro = mpu.get_gyro()?;
        println!("gyro: {}", gyro);
        
        // get gyro data, scaled with sensitivity - averaged across n readings (steps) 
        let gyro = mpu.get_gyro_avg(Steps(5))?;
        println!("gyro avg: {}", gyro);
        
        // get accelerometer data, scaled with sensitivity
        let acc = mpu.get_acc()?;
        println!("acc: {}", acc);
        
        // get accelerometer data, scaled with sensitivity - averages across n readings (steps)
        let acc = mpu.get_acc_avg(Steps(5))?;
        println!("acc avg: {}", acc);		
    }
}

Compile linux example (Raspberry Pi 3B)

Requirements:

$ apt-get install -qq gcc-arm-linux-gnueabihf libc6-armhf-cross libc6-dev-armhf-cross

Rustup:

$ rustup target add armv7-unknown-linux-gnueabihf

To configure the linker use .cargo/config file

cross-compile with

$ cargo build --examples --target=armv7-unknown-linux-gnueabihf

TODO

• init with default settings
• init with custom settings
  • custom sensitivity
  • custom device initialization
• device verification
• basic feature support as described here
• read gyro data
• read acc data
• software calibration
• software measurement variance estimation
• roll, pitch estimation accelerometer only
• read temp data
• rename constants to better match datasheet
• complementary filter for roll, pitch estimate, possible on device?
• sample rate devider with register 25? or timer/clock control with PWR_MGMT_2
  • internal clock, register 108 POWER_MGMT_2, will cycle between sleep mode and waking up to take a single sample of data from active sensors at a rate determined by LP_WAKE_CTRL (page 41-43)
• plotting csv datafor testing? -> See viz branch