Go Down

Topic: Guide to gyro and accelerometer with Arduino including Kalman filtering (Read 325981 times) previous topic - next topic

richardtheboffin

Hi Lauszus,

I've been learning a lot from your code, thanks very much for putting it on here.

I have a board with single axis gyro with 3 axis accelerometer, analogue outputs.

I'm looking to find the Roll angle, so the single gyro board will need to be orientated up on its end so that the Roll axis goes through the middle of the gyro. The accelerometers with then be: X=up/down, Y=left/right, Z=forward/backward.

In this configuration which combination of the X,Y and Z accelerometers should I be using in the code?

I am guessing that it's the X and Y as the Z axis accelerometer is not effected by roll when the board is mounted on it's end

Your code uses X and Z and Y and Z in the arctan function, so I should use X and Y, correct?

Lauszus

@richardtheboffin
It depends on your gyro - some have to be orientated up others don't.
Regarding which axis you should use on your accelerometer: I can't say for sure. It depends on how it's orientated. Either it's X and Z or Y and Z.
Try to modify this code: https://github.com/TKJElectronics/Example-Sketch-for-IMU-including-Kalman-filter/blob/master/IMU6DOF/LPR530AL_%20LY530ALH_%20ADXL335/LPR530AL_%20LY530ALH_%20ADXL335.ino, so it's uses the correct sensitivity values for your setup.
Then afterwards you can easily see which angles should be fused together.
You might also need to invert some of the readings, as I have done here: https://github.com/TKJElectronics/Example-Sketch-for-IMU-including-Kalman-filter/blob/master/IMU6DOF/LPR530AL_%20LY530ALH_%20ADXL335/LPR530AL_%20LY530ALH_%20ADXL335.ino#L51

richardtheboffin

Hi Lauszus,

I have your code running ok with the single gyro and x,y,z accelerometers.

I'll dump the intermediate angle calculation values out and see if I can figure which pair I need to combine with the gyro.

Cheers,

Richard.

richardtheboffin

Hi Lauszus,

All sorted out now. I had everything coded correctly but missed the initial offset from zero for the down axis accelerometer due to gravity.

It works very well and rejects linear movements really well.

Just need to feed the angle into my motor controller now.

Thanks for everything!


tensa

i was able to implement the filter thanks for this guide it was  really helpful...thank you very much
but i dont know how to relate these values to motor's rpm of a quadcopter so i will stabilize, can anybody post links to guides like this?


AntR


i was able to implement the filter thanks for this guide it was  really helpful...thank you very much
but i dont know how to relate these values to motor's rpm of a quadcopter so i will stabilize, can anybody post links to guides like this?




You will need to use a PID controller to give roll,pitch,yaw outputs when the orientation deviates from 0,0,0.  However,  I have used these filters on a quadcopter and they suffer from the problem of linear acceleration affecting the pitch and roll.  The filters are fast enough, smooth and very stable (calculation time was 80micros for twin Kalmans on an Arduino Due) but you'd probably be better off using Mahonys DCM or Madgwicks filter - they're less sensitive to linear acceleration but a bit more prone to noise (vibration from the motors).

You'll find plenty of links via Google for any of this stuff.

dheerajp

i want to work with MPU-6050-Triple Axis Accelerometer Gyro can it be done

dheerajp

i want the same guide for MPU-6050-Triple Axis Accelerometer Gyro

the same think , so my ,i'd like this idea and i would like  someone to share us the GUIDE of IMU 6050 and arduino

Lauszus

Hi,

The setup is very similar just note that if you are using an Arduino running at 5V you might damage the IMU as it's only running at 3.3V, but some breakout boards have a logic level circuit built-in. If not use have to use a logic level converter like this one: https://www.sparkfun.com/products/8745.

To connect the Arduino with the MPU-6050, you need to connect it as it says in the following comment: https://www.sparkfun.com/products/11028#comment-50b3c165ce395f484d000000.

Hopes this helps you out for now, as I havn't got time to write a guide at the moment.

Regards
Lauszus

stunito0o0

Hi Lauszus,
I make a little customisation of your source for digital IMUs so it can work with  Minimu-9 board , hope you don't mind.But there is something wrong here is what i get:
Code: [Select]

....(the board is on the ground and i don't touch it the whole time)
X: 180.00 Y: 180.01
X: 180.05 Y: 180.06
X: 180.22 Y: 180.14
X: 180.51 Y: 180.16
X: 180.82 Y: 180.11
X: 181.07 Y: 180.04
X: 181.42 Y: 179.92
X: 181.79 Y: 179.80
X: 182.22 Y: 179.65
X: 182.72 Y: 179.46
X: 183.23 Y: 179.28
X: 183.85 Y: 179.08
X: 184.54 Y: 178.80
X: 185.26 Y: 178.48
X: 185.99 Y: 178.02
X: 186.60 Y: 177.45
X: 187.54 Y: 176.95
X: 188.54 Y: 176.43
X: 189.64 Y: 175.72
X: 190.82 Y: 175.03
X: 192.13 Y: 174.37
......
X: 280.48 Y: 86.86
X: 280.43 Y: 86.66
X: 280.32 Y: 86.91
X: 280.19 Y: 87.02
X: 280.09 Y: 87.10
X: 279.97 Y: 87.32
X: 279.81 Y: 87.84
X: 279.71 Y: 87.88
X: 279.60 Y: 88.26
X: 279.51 Y: 88.09
X: 279.38 Y: 88.20
X: 279.27 Y: 88.09
X: 279.10 Y: 88.24
X: 278.96 Y: 88.26
X: 278.80 Y: 88.56
...(finally the X stopped growing and the Y stopped getting lower)
X: 278.61 Y: 89.11
X: 278.44 Y: 89.23
X: 278.29 Y: 89.41
X: 278.25 Y: 88.81
X: 278.07 Y: 89.24
X: 277.85 Y: 89.62
X: 277.68 Y: 89.88
X: 277.53 Y: 90.03
X: 277.36 Y: 90.22
X: 277.24 Y: 90.48
X: 277.12 Y: 90.17
X: 276.99 Y: 90.08
X: 276.90 Y: 89.87
X: 276.76 Y: 89.84
X: 276.61 Y: 89.97
X: 276.44 Y: 90.35
X: 276.26 Y: 90.62
X: 276.12 Y: 90.84
X: 276.02 Y: 90.76
X: 275.78 Y: 91.38
X: 275.67 Y: 91.39
X: 275.47 Y: 91.74
X: 275.32 Y: 91.92
X: 275.16 Y: 92.18
X: 275.07 Y: 92.02
X: 274.95 Y: 92.17
X: 274.83 Y: 92.30
X: 274.68 Y: 92.31
X: 274.52 Y: 92.53
X: 274.43 Y: 92.53
X: 274.27 Y: 92.90
X: 274.18 Y: 92.90
X: 274.03 Y: 92.95
..(and all over again)

and here is my source

Code: [Select]
#include <Wire.h>
#include <L3G.h>
#include <LSM303.h>

L3G gyro;
LSM303 compass;
double zeroValue[5] = {-200, 44, 660, -18.5, 52.3}; // Found by experimenting

/* All the angles start at 180 degrees */
double gyroXangle = 180;
double gyroYangle = 180;

double compAngleX = 180;
double compAngleY = 180;

unsigned long timer;
/* Kalman filter variables and constants */
const double Q_angleX = 0.001; // Process noise covariance for the accelerometer - Sw
const double Q_gyroX = 0.003; // Process noise covariance for the gyro - Sw
const double R_angleX = 0.03; // Measurement noise covariance - Sv

double angleX = 180; // The angle output from the Kalman filter
double biasX = 0; // The gyro bias calculated by the Kalman filter
double PX_00 = 0, PX_01 = 0, PX_10 = 0, PX_11 = 0;
double dtX, yX, SX;
double KX_0, KX_1;

double kalmanX(double newAngle, double newRate, double dtime) {
 // KasBot V2 - Kalman filter module - http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1284738418
 // See also http://www.x-firm.com/?page_id=145
 // with slightly modifications by Kristian Lauszus
 // See http://academic.csuohio.edu/simond/courses/eec644/kalman.pdf and
 // http://www.cs.unc.edu/~welch/media/pdf/kalman_intro.pdf for more information
 dtX = dtime / 1000000; // Convert from microseconds to seconds

 // Discrete Kalman filter time update equations - Time Update ("Predict")
 // Update xhat - Project the state ahead
 angleX += dtX * (newRate - biasX);

 // Update estimation error covariance - Project the error covariance ahead
 PX_00 += -dtX * (PX_10 + PX_01) + Q_angleX * dtX;
 PX_01 += -dtX * PX_11;
 PX_10 += -dtX * PX_11;
 PX_11 += +Q_gyroX * dtX;

 // Discrete Kalman filter measurement update equations - Measurement Update ("Correct")
 // Calculate Kalman gain - Compute the Kalman gain
 SX = PX_00 + R_angleX;
 KX_0 = PX_00 / SX;
 KX_1 = PX_10 / SX;

 // Calculate angle and resting rate - Update estimate with measurement zk
 yX = newAngle - angleX;
 angleX += KX_0 * yX;
 biasX += KX_1 * yX;

 // Calculate estimation error covariance - Update the error covariance
 PX_00 -= KX_0 * PX_00;
 PX_01 -= KX_0 * PX_01;
 PX_10 -= KX_1 * PX_00;
 PX_11 -= KX_1 * PX_01;

 return angleX;
}
/* Kalman filter variables and constants */
const double Q_angleY = 0.001; // Process noise covariance for the accelerometer - Sw
const double Q_gyroY = 0.003; // Process noise covariance for the gyro - Sw
const double R_angleY = 0.03; // Measurement noise covariance - Sv

double angleY = 180; // The angle output from the Kalman filter
double biasY = 0; // The gyro bias calculated by the Kalman filter
double PY_00 = 0, PY_01 = 0, PY_10 = 0, PY_11 = 0;
double dtY, yY, SY;
double KY_0, KY_1;

double kalmanY(double newAngle, double newRate, double dtime) {
 // KasBot V2 - Kalman filter module - http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1284738418
 // See also http://www.x-firm.com/?page_id=145
 // with slightly modifications by Kristian Lauszus
 // See http://academic.csuohio.edu/simond/courses/eec644/kalman.pdf and
 // http://www.cs.unc.edu/~welch/media/pdf/kalman_intro.pdf for more information
 dtY = dtime / 1000000; // Convert from microseconds to seconds

 // Discrete Kalman filter time update equations - Time Update ("Predict")
 // Update xhat - Project the state ahead
 angleY += dtY * (newRate - biasY);

 // Update estimation error covariance - Project the error covariance ahead
 PY_00 += -dtY * (PY_10 + PY_01) + Q_angleY * dtY;
 PY_01 += -dtY * PY_11;
 PY_10 += -dtY * PY_11;
 PY_11 += +Q_gyroY * dtY;

 // Discrete Kalman filter measurement update equations - Measurement Update ("Correct")
 // Calculate Kalman gain - Compute the Kalman gain
 SY = PY_00 + R_angleY;
 KY_0 = PY_00 / SY;
 KY_1 = PY_10 / SY;

 // Calculate angle and resting rate - Update estimate with measurement zk
 yY = newAngle - angleY;
 angleY += KY_0 * yY;
 biasY += KY_1 * yY;

 // Calculate estimation error covariance - Update the error covariance
 PY_00 -= KY_0 * PY_00;
 PY_01 -= KY_0 * PY_01;
 PY_10 -= KY_1 * PY_00;
 PY_11 -= KY_1 * PY_01;

 return angleY;
}
int readGyroX(void) {
 int data;
 data =gyro.g.x;

 return data;
}
int readGyroY(void) {
 int data;
 data = gyro.g.y;

 return data;
}
double getXangle() {
 double accXval = (double)readAccX()-zeroValue[0];
 double accZval = (double)readAccZ()-zeroValue[2];
 double angle = (atan2(accXval,accZval)+PI)*RAD_TO_DEG;
 return angle;
}
double getYangle() {
 double accYval = (double)readAccY()-zeroValue[1];
 double accZval = (double)readAccZ()-zeroValue[2];
 double angle = (atan2(accYval,accZval)+PI)*RAD_TO_DEG;
 return angle;
}
int readAccX(void) {
 int data;
 data = compass.a.x;
 return data;
}
int readAccY(void) {
 int data;
 data = compass.a.y;
 return data;
}
int readAccZ(void) {
 int data;
 data =compass.a.z;
 return data;

}


void setup() {
 Serial.begin(9600);
 Wire.begin();
 
 if (!gyro.init())
 {
   Serial.println("Failed to autodetect gyro type!");
   while (1);
 }

 gyro.enableDefault();
  compass.init();
 compass.enableDefault();
   timer = micros();
}

void loop() {
 gyro.read();
compass.read();

 double gyroXrate = (((double)readGyroX()-zeroValue[3])/14.375);
 gyroXangle += gyroXrate*((double)(micros()-timer)/1000000); // Without any filter
 
 double gyroYrate = -(((double)readGyroY()-zeroValue[4])/14.375);
 gyroYangle += gyroYrate*((double)(micros()-timer)/1000000); // Without any filter
 
 double accXangle = getXangle();
 double accYangle = getYangle();


 compAngleX = (0.93*(compAngleX+(gyroXrate*(double)(micros()-timer)/1000000)))+(0.07*accXangle);
 compAngleY = (0.93*(compAngleY+(gyroYrate*(double)(micros()-timer)/1000000)))+(0.07*accYangle);
 
 double xAngle = kalmanX(accXangle, gyroXrate, (double)(micros() - timer)); // calculate the angle using a Kalman filter
 double yAngle = kalmanY(accYangle, gyroYrate, (double)(micros() - timer)); // calculate the angle using a Kalman filter
 
 timer = micros();
 

 Serial.print("X: ");Serial.print(xAngle);Serial.print("\t");
 Serial.print("Y: ");Serial.print(yAngle);Serial.print("\t");
 
 Serial.print("\n");
 
 delay(10);


}

Could you please take a look of my source and tell me what I am doing wrong.I want to thank you in advance for your time.
Stanislav.

Lauszus

@stunito0o0
It looks like you are reading the magnetometer (compass) and not the accelerometer.
Btw the newest version of the code can be found here: https://github.com/TKJElectronics/KalmanFilter.

stunito0o0

#403
Mar 14, 2013, 10:31 am Last Edit: Mar 14, 2013, 10:33 am by stunito0o0 Reason: 1

@stunito0o0
It looks like you are reading the magnetometer (compass) and not the accelerometer.
Btw the newest version of the code can be found here: https://github.com/TKJElectronics/KalmanFilter.

Hi again, :)
I thought too that it might be the compass data I am reading so i opened the example for the "LSM303.h" libraly and here is the source:
Code: [Select]
#include <Wire.h>
#include <LSM303.h>

LSM303 compass;

void setup() {
 Serial.begin(9600);
 Wire.begin();
 compass.init();
 compass.enableDefault();
}

void loop() {
 compass.read();

 Serial.print("A ");
 Serial.print("X: ");
 Serial.print((int)compass.a.x);
 Serial.print(" Y: ");
 Serial.print((int)compass.a.y);
 Serial.print(" Z: ");
 Serial.print((int)compass.a.z);

 Serial.print(" M ");  
 Serial.print("X: ");
 Serial.print((int)compass.m.x);
 Serial.print(" Y: ");
 Serial.print((int)compass.m.y);
 Serial.print(" Z: ");
 Serial.println((int)compass.m.z);

 
 delay(100);
}

According to the example  "compass.a.x","compass.a.y" and "compass.a.z" are for reading the acc data.I will really appreciate if you take another look of my code.I think that when i was customising your source i might have deleted something important.Thank you again for your time. :)
P.S. Now i rememberd that i deleted part of the source for getting the data of acc and gyro.For example "readAccX" function  was something like this:
Code: [Select]
int readAccX(void) {
 int data;
 data = i2cRead(adxlAddress, 0x32);
 data |= i2cRead(adxlAddress,0x33)<<8;
 return data;
}

and i didn't understand what  "data |= i2cRead(adxlAddress,0x33)<<8;" is for so i deleted it.
now it looks like this:
Code: [Select]
int readAccX(void) {
 int data;
 data = compass.a.x;
 return data;
}

Lauszus

Sorry I was in a hurry when I wrote the first reply.
First of all I think it will be easier for you to if you modify this example: https://github.com/pololu/LSM303/blob/master/LSM303/examples/Serial/Serial.ino to your needs.

First of all you should NOT subtract the zeroValues, as they are specific for the IMU used in the example.
I think you can use your values directly, so you don't even have to subtract any zeroValue.
Take a look at this example: https://github.com/TKJElectronics/Example-Sketch-for-IMU-including-Kalman-filter/blob/master/IMU6DOF/MPU6050/MPU6050.ino

Also you assume that the gyro has the same sensitivity as the one used in the example which is most likely not the case (take a look at your datasheet)!

Go Up