Magnetometer calibration and heading (yaw) #3
Comments
|
Panos, The sketch has no calibration for the magnetometer except the manual calibration I might have done for my particular MPU9150. I suggest at least you record the min/max value on all three axes and subtract the average so as to re-center the mag response surface. This is all I did. If you suggest something more sophisticated and automatic and get it to work well I would love to incorporate it into my MPU9x50 and LSM9DS0 sketches. Kris From: ΠΜ Hi Kris, I have got it to work with the https://www.sparkfun.com/products/11486, but the yaw values don't range 360 degrees as I rotate the sensors on its axis (rotating around itself, flat on the table mag-z) and there is not good response, so I may turn it 90 degrees North and it only registers 10º change. The other two pitch and roll seem to work fine. How good is the calibration? Is it worth implementing something like this? http://www.camelsoftware.com/firetail/blog/uavs/3-axis-magnetometer-calibration-a-simple-technique-for-hard-soft-errors/ Many thanks, — |
|
Kris, will have a thought on that. The link I said has a good solution to remove both hard-iron and soft-iron bias. Also quite quick to implement. My question holds. the yaw (headings) gives neither a 360º nor a -180º to 180º output but rather a almost random result. I am using the mpu9150 sparkfun SM11486. Also I am confused with which sensor alignment you finally choose as reference the acc or the mag? So which xyz should I use to find north etc? Finally I had a look on the MPU6050_9Axis_MotionApps41.h and it seems it also has functions to calculate the euler quaternions etc (e.g. MPUgetyawptichroll) - but you are not using those, correct? If that's the case why? Cheers, |
|
Hi Panos, I am using the x,y,z of the accel/gyro as reference. If you do not calibrate the magnetometer the Euler values of yaw, pitch, and roll can be quite non-sensical. I wrote my own AHRS routines based on open-source Madgwick and Mahoney sensor fusion algorithms. I don’t like the way Invensense programs are structured nor do I like their dependence on a binary blob for the DMP, which they also depend on. Kris From: ΠΜ Kris, will have a thought on that. The link I said has a good solution to remove both hard-iron and soft-iron bias. Also quite quick to implement. My question holds. the yaw (headings) gives neither a 360º nor a -180º to 180º output but rather a almost random result. I am using the mpu9150 sparkfun SM11486. Also I am confused with which sensor alignment you finally choose as reference the acc or the mag? So which xyz should I use to find north etc? Finally I had a look on the MPU6050_9Axis_MotionApps41.h and it seems it also has functions to calculate the euler quaternions etc (e.g. MPUgetyawptichroll) - but you are not using those, correct? If that's the case why? Cheers, — |
|
Hello Kris, After implementing Madwick filter, we have the quaternion representing orientation of MPU9150 in global co-ordinate frame. Now if I want to represent the quaternion in the body co-ordinate frame what should I do ? Thanks in advance, |
|
Hi Akhil, I am not sure what you want to do. The quaternion represents the sensor's Kris -----Original Message----- Hello Kris, After implementing Madwick filter, we have the quaternion representing Thanks in advance, Reply to this email directly or view it on GitHub |
|
Hello, Thanks for the quick reply. |
|
If you have absolute orientation for each sensor, can you not rather simply calculate the relative angle between, say, the two IMU's x-axes? You have two orientation vectors wrt the same global absloute reference; hwy can you not simply compare these two orientations directly? Why do you need a transform? |
|
Yes, we will get a relative orientation of one segment with other in this way. But the problem is i don't know the value of a at time = 0. This can be known only when you keep your body in a fixed pose. For that, initially the segment under measurement is calibrated. For calibration - (R)GB = (R)SB x (R)GS We can do all these algebra in quaterion form and is mathematically and computationally useful. For me, I am nor sure how do we represent (R)GB (Is it a (I)3x3 matrix at rest) ? I don't know. |
|
I am not sure what you are asking. If you know each sensors absolute orientation (i.e., quaternion) wrt a fixed global reference like True North, don't you know everything you need to know? |
|
Hello, Using rotation matrix I know that (R)GB = (R)SB x (R)GS. I have tried doing qSB = qGB x q*GS (where * corresponds to Quaternion conjugate and x quaternion multiplication). Here qGB = [0.707, 0, 0, -0.707] and which results in singularities and I don't know how to solve this. Hope this explanation is fine. I am really sorry if I am not able to convey what I want. Thanks for your patience Kris. |
|
Hi Kris, |
|
Here is a mag calibration routine for the MPU9250 which can be adapted for the MPU9150: void magcalMPU9250(float * dest1) |
|
Thanks so much for answering so fast ! Still working with the code, probably got another few questions popping up, On 10 April 2015 at 11:13, Kris Winer notifications@github.com wrote:
|
|
I write the accel and gyro biases to user bias registers. These bias values Kris -----Original Message----- Thanks so much for answering so fast ! Still working with the code, probably got another few questions popping up, On 10 April 2015 at 11:13, Kris Winer notifications@github.com wrote:
Reply to this email directly or view it on GitHub |
|
Ok, got that. One more thing, how do define the user-environmental x,y,z On 10 April 2015 at 14:24, Kris Winer notifications@github.com wrote:
|
|
Do you mean the mag biases, as below or what? The real mag response surface is generally off-center and ellipsoidal, not -----Original Message----- Ok, got that. One more thing, how do define the user-environmental x,y,z On 10 April 2015 at 14:24, Kris Winer notifications@github.com wrote:
Reply to this email directly or view it on GitHub |
|
Hello kris, heading_dir = 90 - ( atan ( mx / my ) ) * ( 180 / PI ) - 23.5; //here, 23.5 is a declination angle. So,any idea by which i can find that which direction is vertically down?. kindly reply. |
|
Why don't you want to use 9 DoF sensor fusion for heading estimation? It is a much more accurate solution. Your approach can only work in some limited circumstances. And then not very well. |
|
I have no idea about 9 DoF sensor.Can you explain me,how can i use and how can access and how it is useful?. |
|
Actually,I have using IMU/mpu-9250 to detect position of device. In our project,we will attach our device by OBD connector. But problem is critical for that.Because the OBD connector has attached in different types for different device.So,magnetic direction will changes. |
|
9DoF sensor fusion like used here:
https://github.com/kriswiner/MPU-9250/blob/master/MPU9250_MS5637_AHRS_t3.ino
…On Fri, May 12, 2017 at 10:13 PM, jenextech ***@***.***> wrote:
Actually,I have using IMU/mpu-9250 to detect position of device. In our
project,we will attach our device by OBD connector.
But problem is critical for that.Because the OBD connector has attached in
different types for different device.So,magnetic direction will changes.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1quUjCiWX_Y1eueysRfdDtcgM9LBmks5r5TwMgaJpZM4CuKSq>
.
|
|
so,its for find yaw,pitch and roll...right? |
|
I refer code which you told. |
|
You need to calibrate your sensors.
…On Thu, Jun 1, 2017 at 4:43 AM, jenextech ***@***.***> wrote:
I refer code which you told.
Actually ,i want to measure direction and its work.But problem is
that,device installation-direction is not fix.So,device movement direction
will change(false).So,how can i use yaw,pitch and roll reading to measure
device's movement with direction?.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qryIbDY6BTNgnWMZGkjyUcOITf-gks5r_qPPgaJpZM4CuKSq>
.
|
|
see this image.... now,still issues occur.....see as below....
now if 2 cars stay in opposite direction and both moving clockwise.....then it shows same direction sequence...( NW -> NE -> SE -> SW ) ..... In our condition....car_direction not fixed....device_direction not fixed....so this type of issue happen...... so,how can i use magnetometer data....to detect car movement with direction.... |
|
i can get completely result from IMU to detect car movement.But it will right when IMU put at fixed position.But in our condition if many cars stay in north direction , but IMU may be attached in car with different type of position.So,if all cars moving in one direction.But it shows different direction.Because of device attached in different style in cars. |
|
How do you calibrate the sensors?
…On Fri, Jun 2, 2017 at 11:40 PM, jenextech ***@***.***> wrote:
i can get completely result from IMU to detect car movement.But it will
right when IMU put at fixed position.But in our condition if many cars stay
in north direction , but IMU may be attached in car with different type of
position.So,if all cars moving in one direction.But it shows different
direction.Because of device attached in different style in cars.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qgg0SGjiZTy84Zz8e-fj5gljBBm5ks5sAP_hgaJpZM4CuKSq>
.
|
|
i show you my code , wherever i have used your code ,i just call that function(not write definition) initMPU9250(); // its your function //see the definition of the above function // while(1) //useful to see data on the terminal... } initAK8963(); // its also your function readGyroData_jt(gyroCount); // if( ((gx < -2.6) || (gx > 2.6)) || ((gy < -1.2) || (gy > 1.2)) || ((gz < -3.0) || (gz > 3.0)) ) readMagData_jt(magCount); // if(counter == 0) switch(direction_section) // if( counter == 1 ) // if( counter == 1 ) //// if( (mx>0) && (my>0) ) // if( counter == 1 ) // if( car_moving == TRUE ) |
|
I don;t see where you are calibrating the sensors.
…On Sun, Jun 4, 2017 at 10:11 PM, jenextech ***@***.***> wrote:
i show you my code , wherever i have used your code ,i just call that
function(not write definition)
i have initialize imu and do configure as your code...
then i have get raw data of accelerometer and check which axis is pointing
to the vertically down.
on basis of it,heading direction will fixed....
i have made changes in code.so you can see some are commented....
initMPU9250(); // its your function
initial_compensation(); // in this function i have checking that which
axis is pointing vertically down.
//see the definition of the above function
void initial_compensation()
{
static int ini_counter,accelCount[3];
do
{
readByte_jt(MPU9250_ADDRESS,INT_STATUS,data);
if(data[0] & 0x01)
{
readAccelData_jt(accelCount);
Bias[0] += (float)accelCount[0];
Bias[1] += (float)accelCount[1];
Bias[2] += (float)accelCount[2];
ini_counter++;
}
}while(ini_counter<10);
Bias[0] /=10;
Bias[1] /=10;
Bias[2] /=10;
/*NOTE :-
* Below code is useful to find that which accelero/gyro meter's axis ( X,Y or Z ) is in vertical down direction.
* It is necessary.Because formula for finding device location is depend on axis.So,for different vertical down direction-axis,axis in formula will be change.
*/
mag_dir_x = (Bias[0]/16384) * 9.8;
mag_dir_y = (Bias[1]/16384) * 9.8;
mag_dir_z = (Bias[2]/16384) * 9.8;
// while(1) //useful to see data on the terminal...
// {
//
// uart_transmit_string_jt("mag_dir_x = ",12);
// uart_transmit_float_jt(mag_dir_x);
// uart_transmit_char_jt(' ');
// uart_transmit_float_jt(Bias[0]);
// uart_transmit_char_jt('\n');
//
// uart_transmit_string_jt("mag_dir_y = ",12);
// uart_transmit_float_jt(mag_dir_y);
// uart_transmit_char_jt(' ');
// uart_transmit_float_jt(Bias[1]);
// uart_transmit_char_jt('\n');
//
// uart_transmit_string_jt("mag_dir_z = ",12);
// uart_transmit_float_jt(mag_dir_z);
// uart_transmit_char_jt(' ');
// uart_transmit_float_jt(Bias[2]);
// uart_transmit_char_jt('\n');
//
// }
if( (mag_dir_x > 8) || (mag_dir_x < (-8)) )
{
// vertical_down_direction is accelero/gyro meter x-axis.
if( mag_dir_x > 8 )
direction_section = 1; // +ve X-axis
else
direction_section = 2; // -ve X-axis
}
else if( (mag_dir_y > 8) || (mag_dir_y < (-8)) )
{
// vertical_down_dirextion is accelero/gyro meter y-axis.
if( mag_dir_y > 8 )
direction_section = 3; // +ve Y-axis
else
direction_section = 4; // -ve Y-axis
}
else if( (mag_dir_z > 8) || (mag_dir_z < (-8)) )
{
// vertical_down_dirextion is accelero/gyro meter z-axis.
if( mag_dir_z > 8 )
direction_section = 5; // +ve Z-axis
else
direction_section = 6; // -ve Z-axis
}
}
initAK8963(); // its also your function
readGyroData_jt(gyroCount);
gx = (float)gyroCount[0]/(float)131;// - gyroBias[0]; // get actual gyro
value, this depends on scale being set
gy = (float)gyroCount[1]/(float)131;// - gyroBias[1];
gz = (float)gyroCount[2]/(float)131;// - gyroBias[2];
// if( ((gx < -2.6) || (gx > 2.6)) || ((gy < -1.2) || (gy > 1.2)) || ((gz
< -3.0) || (gz > 3.0)) )
// if( ((gx < -10.0) || (gx > 10.0)) || ((gy < -10.0) || (gy > 10.0)) ||
((gz < -5.0) || (gz > 5.0)) )
if( ((gx < -5.0) || (gx > 5.0)) || ((gy < -5.0) || (gy > 5.0)) || ((gz < -5.0) || (gz > 5.0)) )
car_moving = TRUE;
else
car_moving = FALSE;
readAccelData_jt(accelCount);
accelCount[0] -=(int) Bias[0];
accelCount[1] -=(int) Bias[1];
accelCount[2] -=(int) Bias[2];
ax = (float)accelCount[0]/(float)16384;//((float)2/32768);// - accelBias[0]; // get actual g value, this depends on scale being set
ay = (float)accelCount[1]/(float)16384;// - accelBias[1];
az = (float)accelCount[2]/(float)16384;// - accelBias[2];
ax *= 9.8;
ay *= 9.8;
az *= 9.8;
if(ax < 0.3 && ax > -0.3)
ax=0;
if(ay < 0.3 && ay > -0.3)
ay=0;
if(az < 0.3 && az > -0.3)
az=0;
readMagData_jt(magCount);
// Calculate the magnetometer values in milliGauss
// Include factory calibration per data sheet and user environmental corrections
mx = (float)magCount[0]*(1/0.6);//*mRes*magCalibration[0] - magbias_1[0]; // get actual magnetometer value, this depends on scale being set
my = (float)magCount[1]*(1/0.6);//*mRes*magCalibration[1] - magbias_1[1];
mz = (float)magCount[2]*(1/0.6);//*mRes*magCalibration[2] - magbias_1[2];
switch_to_acce_gyro_meter_jt();
// if(counter == 0)
// {
// switch(direction_section)
// {
//
// case 1: side_dir = mx;//mx;
// down_dir = mz;//mz;
// break;
//
// case 2: side_dir = mx;//mz; //if here changing
sign(-ve),nissan_car/swift_car detect same in direction,but angle may be
different.
// down_dir = -mz;//mx; //if here changing sign(-ve),nissan_car/swift_car
detect same in direction,but angle may be different.
// break;
//
// case 3:
// if( ((my > 150)&&(mx<250)) && ((mz > -50)&&(mz<160)) )
// {
// counter = 1;
// break;
// }
//
// else if( ((my > -250)&&(mx < -100)) && ((mz>150)&&(mz<240)) )
// {
// counter = 2;
// break;
// }
//
// else if( ((my > -380)&&(mx < -300)) && ((mz > -230)&&(mz < -90)) )
// {
// counter = 3;
// break;
// }
//
// else if( ((my > -140)&&(mx<170)) && ((mz > -240)&&(mz < -400)) )
// {
// counter = 4;
// break;
// }
//
// case 4:
// if( ((my > -430)&&(mx < -280)) && ((mz > 30)&&(mz<210)) )
// {
// counter = 1;
// break;
// }
//
// else if( ((my > -110)&&(mx < 70)) && ((mz>310)&&(mz<210)) )
// {
// counter = 2;
// break;
// }
//
// else if( ((my > 120)&&(mx < 200)) && ((mz > -210)&&(mz < -70)) )
// {
// counter = 3;
// break;
// }
//
// else if( ((my > -330)&&(mx < -150)) && ((mz > -330)&&(mz < -200)) )
// {
// counter = 4;
// break;
// }
//
//
// case 5:
// if( ((mx > -100)&&(mx<150)) && ((my>100)&&(my<250)) )
// {
// counter = 1;
// break;
// }
//
// else if( ((mx > -230)&&(mx < -120)) && ((my > -390)&&(my < -350)) )
// {
// counter = 2;
// break;
// }
//
// else if( ((mx > -400)&&(mx < -430)) && ((my > -80)&&(my<80)) )
// {
// counter = 3;
// break;
// }
//
// else if( ((mx > 180)&&(mx<210)) && ((my > -220)&&(my < -140)) )
// {
// counter = 4;
// break;
// }
//
// case 6:
// if( ((mx > -60)&&(mx < 80)) && ((my > -380)&&(my < -290)) )
// {
// counter = 1;
// break;
// }
//
// else if( ((mx > 140)&&(mx < 190)) && ((my > -100)&&(my < 60)) )
// {
// counter = 2;
// break;
// }
//
// else if( ((mx > -260)&&(mx < -110)) && ((my > 180)&&(my<210)) )
// {
// counter = 3;
// break;
// }
//
// else if( ((mx > -420)&&(mx < -360)) && ((my > -270)&&(my < -90)) )
// {
// counter = 4;
// break;
// }
//
// default:
// break;
// }
// }
switch(direction_section)
{
case 1: side_dir = mx;//mx;
down_dir = mz;//mz;
break;
case 2: side_dir = mx;//mz; //if here changing sign(-ve),nissan_car/swift_car detect same in direction,but angle may be different.
down_dir = -mz;//mx; //if here changing sign(-ve),nissan_car/swift_car detect same in direction,but angle may be different.
break;
case 3:
// if( counter == 1 )
// {
// side_dir = mz; //mx;
// down_dir = my; //my;
//
// break;
// }
//
// else if( counter == 2 )
// {
// side_dir = -my; //mx;
// down_dir = mz; //my;
//
// break;
// }
//
// else if( counter == 3 )
// {
// side_dir = -mz; //mx;
// down_dir = -my; //my;
//
// break;
// }
//
// else if( counter == 4 )
// {
// side_dir = my; //mx;
// down_dir = -mz; //my;
//
// break;
// }
side_dir = my;//mz;
down_dir = -mz;//my;
break;
case 4:
// if( counter == 1 )
// {
// side_dir = mz; //mx;
// down_dir = -my; //my;
//
// break;
// }
//
// else if( counter == 2 )
// {
// side_dir = my; //mx;
// down_dir = mz; //my;
//
// break;
// }
//
// else if( counter == 3 )
// {
// side_dir = -mz; //mx;
// down_dir = my; //my;
//
// break;
// }
//
// else if( counter == 4 )
// {
// side_dir = -my; //mx;
// down_dir = -mz; //my;
//
// break;
// }
side_dir = my;//my; //if here changing sign(-ve),nissan_car/swift_car detect same in direction,but angle may be different.
down_dir = mz;//mz; //if here changing sign(-ve),nissan_car/swift_car detect same in direction,but angle may be different.
break;
case 5:
//// if( (mx>0) && (my>0) )
// if( counter == 1 )
// {
// side_dir = mx; //mx;
// down_dir = my; //my;
//
// break;
// }
////
//// else if( (mx<0) && (my<0) )
// else if( counter == 2 )
// {
// side_dir = -mx; //mx;
// down_dir = -my; //my;
//
// break;
// }
//// else if( (mx<0) && (my>0) )
// else if( counter == 3 )
// {
// side_dir = my; //mx;
// down_dir = -mx; //my;
//
// break;
// }
////
//// else if( (mx>0) && (my<0) )
// else if( counter == 4 )
// {
// side_dir = -my; //mx;
// down_dir = mx; //my;
//
// break;
// }
side_dir = mx; //mx;
down_dir = my; //my;
break;
case 6:
// if( counter == 1 )
// {
// side_dir = mx; //mx;
// down_dir = -my; //my;
//
// break;
// }
//
// else if( counter == 2 )
// {
// side_dir = my; //mx;
// down_dir = mx; //my;
//
// break;
// }
//
// else if( counter == 3 )
// {
// side_dir = -mx; //mx;
// down_dir = my; //my;
//
// break;
// }
//
// else if( counter == 4 )
// {
// side_dir = -my; //mx;
// down_dir = -mx; //my;
//
// break;
// }
side_dir = mx;//my; //if here changing sign(-ve),nissan_car/swift_car detect same in direction,but angle may be different.
down_dir = -my;//mx; //if here changing sign(-ve),nissan_car/swift_car detect same in direction,but angle may be different.
break;
default:
break;
}
/*NOTE :-
* here, 23.5 is a declination angle,it is changing at every day(means it is difference for all 365 days).
*
* "atan" is the function implemented by programming languages such as C++ for arctan,
* and returns values (in radians) in the range: -PI/2 <= atan(x) <= PI/2.
*
* if you want these values in degrees, you must calculate atan(x*180/PI).
*/
if(down_dir == 0 && side_dir < 0)
heading_dir = 180 - 23.5;
else if(down_dir == 0 && side_dir > 0)
heading_dir = 360 - 23.5;
else if(down_dir > 0)
{
heading_dir = 90 - (atan(side_dir/down_dir))*(180/PI) - 23.5;
if(heading_dir < 0)
heading_dir += 360;
}
else if(down_dir < 0)
{
heading_dir = 270 - (atan(side_dir/down_dir))*(180/PI) - 23.5;
if(heading_dir > 360)
heading_dir -= 360;
}
char data_1[3]={0};
if(heading_dir == 0)
data_1[0]='N';
else if(heading_dir == 180)
data_1[0]='S';
else if(heading_dir > 0 && heading_dir < 90)
{
data_1[0]='N';
data_1[1]='W';
}
else if(heading_dir > 90 && heading_dir < 180)
{
data_1[0]='S';
data_1[1]='W';
}
else if(heading_dir > 180 && heading_dir < 270)
{
data_1[0]='S';
data_1[1]='E';
}
else if(heading_dir > 270 && heading_dir < 360)
{
data_1[0]='N';
data_1[1]='E';
}
else
;
heading_dir = 360 - heading_dir;
find_position_jt();
dx = (float)distance*cos(heading_dir);
dy = (float)distance*sin(heading_dir);
delta_longi = (float)dx/(111320 * cos(LAT));
delta_lat = (float)dy / 110540;
F_longi = LONGI +(float)delta_longi;
F_lat = LAT +(float) delta_lat;
// if( car_moving == TRUE )
{
sprintf(data_sensor1,"Heading Direction= %f %s\n",heading_dir, data_1);
uart_transmit_string_jt(data_sensor1,strlen(data_sensor1));
sprintf(data_sensor1,"vx = %f vy = %f vz = %f\n",vx,vy,vz);
uart_transmit_string_jt(data_sensor1,strlen(data_sensor1));
sprintf(data_sensor1,"px = %f py = %f pz = %f\nTime = %fsecs\n distance =
%f \n lat = %f long = %f",px,py,pz,t, distance,F_lat,F_longi);
uart_transmit_string_jt(data_sensor1,strlen(data_sensor1));
uart_transmit_string_jt("\n\n",2);
}
// else
// uart_transmit_string_jt("CAR IS STATIONARY\n",18);
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qqHshnVnF5UijqPwC7e1GyKVJ7faks5sA44CgaJpZM4CuKSq>
.
|
|
for gyrometer.... readGyroData_jt(gyroCount); // if( ((gx < -2.6) || (gx > 2.6)) || ((gy < -1.2) || (gy > 1.2)) || ((gz < -3.0) || (gz > 3.0)) ) for accelerometer.... if(ax < 0.3 && ax > -0.3) is it?.... |
|
And where do gyroBias and accelBias come from?
And magnetometer calibration?
…On Sun, Jun 4, 2017 at 11:19 PM, jenextech ***@***.***> wrote:
for gyrometer....
readGyroData_jt(gyroCount);
gx = (float)gyroCount[0]/(float)131;// - gyroBias[0]; // get actual gyro
value, this depends on scale being set
gy = (float)gyroCount[1]/(float)131;// - gyroBias[1];
gz = (float)gyroCount[2]/(float)131;// - gyroBias[2];
// if( ((gx < -2.6) || (gx > 2.6)) || ((gy < -1.2) || (gy > 1.2)) || ((gz
< -3.0) || (gz > 3.0)) )
// if( ((gx < -10.0) || (gx > 10.0)) || ((gy < -10.0) || (gy > 10.0)) ||
((gz < -5.0) || (gz > 5.0)) )
if( ((gx < -5.0) || (gx > 5.0)) || ((gy < -5.0) || (gy > 5.0)) || ((gz < -5.0) || (gz > 5.0)) )
car_moving = TRUE;
else
car_moving = FALSE;
for accelerometer....
readAccelData_jt(accelCount);
accelCount[0] -=(int) Bias[0];
accelCount[1] -=(int) Bias[1];
accelCount[2] -=(int) Bias[2];
ax = (float)accelCount[0]/(float)16384;//((float)2/32768);// - accelBias[0]; // get actual g value, this depends on scale being set
ay = (float)accelCount[1]/(float)16384;// - accelBias[1];
az = (float)accelCount[2]/(float)16384;// - accelBias[2];
ax *= 9.8;
ay *= 9.8;
az *= 9.8;
if(ax < 0.3 && ax > -0.3)
ax=0;
if(ay < 0.3 && ay > -0.3)
ay=0;
if(az < 0.3 && az > -0.3)
az=0;
is it?....
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qh0bAGBIZ18pxlu1ZT-_ZDo2w-0Bks5sA54MgaJpZM4CuKSq>
.
|
|
do }while(ini_counter<10); Bias[0] /=10; |
|
Hello Kris Thanks for the great work with your lib. I could really use your input. I have read so many posts, browsed through different examples, tried different things, Madgwick/Mahony updates, +mz, -mz, different calibration methods etc. but nothing yields a usable result for my purpose. Goal and setup:
I want to use the MPU9250 and compensation to get a north heading, so the I can get a north heading compared to the orientation of the MPU9250. Which axis to use is not important, since I can mount the MPU9250 however I like for my project. I would really appreciate a complete code example of how to set up, how to calibrate and how to use orientation compensation (when MPU9250 is rotated around x, y, z) to get heading to north (0-360 or -180-180). Please do not just link to other github libs, since I have already tried thess, and tried to implement calibration, magBias etc. and nothing yields anything usable. I have the code up and running, and reading etc. So not a I2C ussue or wiring usse. But yaw makes no sense and is not the same when rorating back to same position over and over again. Hope to get some input, so I can get my gadget geocache out in the field, for many geocacher to enjoy! Thanks in advance to anybody giving input |
|
Some comments:
I will assume your code is correct and that your are properly calibrating
your sensors.
1) the purple CJMU boards from China are absolute crap and I would never
use them; the circuit design is wrong.
2) you need to run the accel and especially gyro at 1 kHz to get best
results
3) you need to be able to run the fusion algorithm at 10 x or so the sample
rate, so a 16 MHz AVR is not going to cut it. Try a Cortex M4F like this
<https://www.tindie.com/products/TleraCorp/ladybug-stm32l432-development-board/>
one.
4) You want a complete code example but you don't want me to link to a
github repository??? This
<https://github.com/kriswiner/Ladybug/blob/master/MPU9250_MS5637_BasicAHRS2_Ladybug.ino>
code example is complete and gives very good results for me using the MCU I
linked to.
I suspect your basic problem (apart from the crap MPU9250) is your MCU is
too pokey. If you want to use a pokey AVR MCU look into this
<https://www.tindie.com/products/onehorse/ultimate-sensor-fusion-solution/>.
It works with the pokiest MCUs.
…On Sat, Jun 9, 2018 at 6:12 AM, Gubbifisken ***@***.***> wrote:
Hello Kris
Thanks for the great work with your lib. I could really use your input.
I have read so many posts, browsed through different examples, tried
different things, Madgwick/Mahony updates, +mz, -mz, different calibration
methods etc. but nothing yields a usable result for my purpose.
Goal and setup:
I am using
- MPU9250 (https://www.banggood.com/MPU9250-Integrated-9DOF-9-
Axis-Attitude-Accelerometer-Gyro-Compass-Magnetic-Field-
Sensor-For-Arduino-p-1101005.html?rmmds=myorder
<https://www.banggood.com/MPU9250-Integrated-9DOF-9-Axis-Attitude-Accelerometer-Gyro-Compass-Magnetic-Field-Sensor-For-Arduino-p-1101005.html?rmmds=myorder>
)
- Arduino 2560
- Arduino studio
- Etc.
I want to use the MPU9250 and compensation to get a north heading, so the
I can get a north heading compared to the orientation of the MPU9250. Which
axis to use is not important, since I can mount the MPU9250 however I like
for my project.
I would really appreciate a complete code example of how to set up, how to
calibrate and how to use orientation compensation (when MPU9250 is rotated
around x, y, z) to get heading to north (0-360 or -180-180). Please do not
just link to other github libs, since I have already tried thess, and tried
to implement calibration, magBias etc. and nothing yields anything usable.
I have the code up and running, and reading etc. So not a I2C ussue or
wiring usse. But yaw makes no sense and is not the same when rorating back
to same position over and over again.
Hope to get some input, so I can get my gadget geocache out in the field,
for many geocacher to enjoy!
Thanks in advance to anybody giving input
Gibbifisken
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qhsEa4gweHUpz3XKx0pAVfOcHTrMks5t68nOgaJpZM4CuKSq>
.
|
|
Hello Kris I will look into the links you send, and look at upgrading MCU to F4. I wonder why the Chinese products are that bad. I have a 6050 that works great for my gyro project. I also fly quadcopter with parts from China, and they works well. And here we are talking about a quad spinning with 1200 degrees/s very smooth. Do you have a suggestion where to get whatever MPU in EU that it workth buying. I need one that can farely give me north heading (making kind of digital compass). A few degrees of is fine and not a problem Gubbifisken |
|
I'd start with the upgrade to your MCU.
…On Sat, Jun 9, 2018 at 10:49 AM, Gubbifisken ***@***.***> wrote:
Hello Kris
Thansk for the super quick reply. What I meant with not github link was
that the ones I already have have inconsistencies between them (complete
repo but also comments online). An example is that many write about
calibration. Hence I wonder why the calibration function (min, maxand avg
bias) is not just a part of the libs I have found, and then commeted out
(with a comment that one needs to use for finfing specific HW bias for
their setup).
I will look into the links you send, and look at upgrading MCU to F4.
I wonder why the Chinese products are that bad. I have a 6050 that works
great for my gyro project. I also fly quadcopter with parts from China, and
they works well. And here we are talking about a quad spinning with 1200
degrees/s very smooth.
Do you have a suggestion where to get whatever MPU in EU that it workth
buying. I need one that can farely give me north heading (making kind of
digital compass). A few degrees of is fine and not a problem
Gubbifisken
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qsziMgqYDI7haTn95qJxjI0RvBT4ks5t7AqagaJpZM4CuKSq>
.
|
|
Hello, I seem to have the same issue... I have been running the calibration routine that was posted above, but I get rather inconsistent results from it and, when the values are pasted into the code, I still get nonsensical yaw results. |
|
You are using an MPU9150?
Do you have the sensor values arranged as NED when input into the fusion
filter?
…On Thu, Sep 6, 2018 at 5:37 AM JabberwockPL ***@***.***> wrote:
Hello,
I seem to have the same issue...
I have been running the calibration routine that was posted above, but I
get rather inconsistent results from it and, when the values are pasted
into the code, I still get nonsensical yaw results.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qgQL7gLoT8i9fJYtB9a8iQGvYMIjks5uYRclgaJpZM4CuKSq>
.
|
|
Yes, I am using MPU9150. I am not sure what you mean by NED - I am running the calibration procedure you have given above and am plugging the results (dest[0] to [3]) to magbias[0] to [3], as that is what the original MPU9250 program seems to be doing. The values I have obtained were -42; 79; 35; although they vary from measurement to measurement. |
|
Are you calculating quaternions?
…On Thu, Sep 6, 2018 at 10:45 AM JabberwockPL ***@***.***> wrote:
Yes, I am using MPU9150.
I am not sure what you mean by NED - I am running the calibration
procedure you have given above and am plugging the results (dest[0] to [3])
to magbias[0] to [3], as that is what the original MPU9250 program seems to
be doing. The values I have obtained were -42; 79; 35; although they vary
from measurement to measurement.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qtTamGUQi8V3UH38K93zvtjoSmHwks5uYV9TgaJpZM4CuKSq>
.
|
|
For now I am just testing your code and the only changes I have made are removing the display references, changing the declination data and putting the magbias values manually. I have run the calibration again and when using the new values it works much better. It is still rather inaccurate for yaw (turning the module about 90 degrees goes from 0 to 160), but I guess additional calibration will help with that. |
|
Don't assume sketches in this repository are correct, or appropriate for
your application.
" I am not sure what you mean by NED"
This means you choose which edge of your board is going to be North. Then
find out which axxel axis point North, then East, and then down. Same for
gyro and mag. The provide filter the sensor data in the following order:
An, Ae, Ad, Gn, Ge, Gd, Mn, Me, Md) Otherwise you will get nonsense.
…On Fri, Sep 7, 2018 at 12:43 AM JabberwockPL ***@***.***> wrote:
For now I am just testing your code and the only changes I have made are
removing the display references, changing the declination data and putting
the magbias values manually.
I have run the calibration again and when using the new values it works
much better. It is still rather inaccurate for yaw (turning the module
about 90 degrees goes from 0 to 160), but I guess additional calibration
will help with that.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qkRLY5SRaEi7c9frf1E48h1Ql0sNks5uYiOXgaJpZM4CuKSq>
.
|
|
Hi Kris, I've seen you mention several times that you should to run the fusion algorithm much more frequently than the sample rate. Can you explain why this is the case? I've been trying to figure out why that is, but so far havn't come up with anything. |
|
Madgwick and Mahony algorithms are similar to steepest descent methods and
require iteration to asymptote to an accurate solution, just that simple....
…On Fri, Nov 2, 2018 at 2:04 PM dsstewart ***@***.***> wrote:
Hi Kris,
I've seen you mention several times that you should to run the fusion
algorithm much more frequently than the sample rate. Can you explain why
this is the case? I've been trying to figure out why that is, but so far
havn't come up with anything.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#3 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AGY1qv73-8qxFdIZ5vQosS8MFyfQgrBYks5urLN6gaJpZM4CuKSq>
.
|
Hi Kris,
Amazing work porting the MARG into a nice arduino sketch!
I have got it to work with the https://www.sparkfun.com/products/11486, but the yaw values don't range 360 degrees as I rotate the sensors on its axis (rotating around itself, flat on the table mag-z) and there is not good response, so I may turn it 90 degrees North and it only registers 10º change. The other two pitch and roll seem to work fine.
How good is the calibration? Is it worth implementing something like this? http://www.camelsoftware.com/firetail/blog/uavs/3-axis-magnetometer-calibration-a-simple-technique-for-hard-soft-errors/
Many thanks,
Panos
The text was updated successfully, but these errors were encountered: