Mirror
/
ardupilot
mirror of https://github.com/ArduPilot/ardupilot
5
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

DCM: got rid of _error_roll_pitch from object state 

this can be a local error variable in common with the yaw code. This
saves 12 bytes in the object.
This commit is contained in:
Andrew Tridgell 2012-03-01 13:48:35 +11:00
parent f399ada86f
commit 4e354a9731
2 changed files with 12 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
libraries/AP_DCM/AP_DCM.cpp
View File

@ -211,7 +211,6 @@ AP_DCM::matrix_reset(bool recover_eulers)
_omega_P = _omega_I;
_omega_integ_corr = _omega_I;
_omega = _omega_I;
_error_roll_pitch = _omega_I;
// if the caller wants us to try to recover to the current
// attitude then calculate the dcm matrix from the current
@ -360,7 +359,7 @@ AP_DCM::drift_correction(void)
float accel_magnitude;
float accel_weight;
float integrator_magnitude;
Vector3f error_yaw;
Vector3f error;
//static float scaled_omega_P[3];
//static float scaled_omega_I[3];
@ -377,16 +376,18 @@ AP_DCM::drift_correction(void)
_health = constrain(_health+(0.02 * (accel_weight - .5)), 0, 1);
// adjust the ground of reference
_error_roll_pitch = _dcm_matrix.c % _accel_vector; // Equation 27 *** sign changed from prev implementation???
error = _dcm_matrix.c % _accel_vector; // Equation 27 *** sign changed from prev implementation???
// error_roll_pitch are in Accel m/s^2 units
// Limit max error_roll_pitch to limit max omega_P and omega_I
_error_roll_pitch.x = constrain(_error_roll_pitch.x, -1.17f, 1.17f);
_error_roll_pitch.y = constrain(_error_roll_pitch.y, -1.17f, 1.17f);
_error_roll_pitch.z = constrain(_error_roll_pitch.z, -1.17f, 1.17f);
float error_norm;
error_norm = error.length();
if (error_norm > 2) {
error *= (2 / error_norm);
}
_omega_P = _error_roll_pitch * (_kp_roll_pitch * accel_weight);
_omega_I += _error_roll_pitch * (_ki_roll_pitch * accel_weight);
_omega_P = error * (_kp_roll_pitch * accel_weight);
_omega_I += error * (_ki_roll_pitch * accel_weight);
//*****YAW***************
@ -452,10 +453,10 @@ AP_DCM::drift_correction(void)
}
}
error_yaw = _dcm_matrix.c * error_course; // Equation 24, Applys the yaw correction to the XYZ rotation of the aircraft, depeding the position.
error = _dcm_matrix.c * error_course; // Equation 24, Applys the yaw correction to the XYZ rotation of the aircraft, depeding the position.
_omega_P += error_yaw * _kp_yaw; // Adding yaw correction to proportional correction vector.
_omega_I += error_yaw * _ki_yaw; // adding yaw correction to integrator correction vector.
_omega_P += error * _kp_yaw; // Adding yaw correction to proportional correction vector.
_omega_I += error * _ki_yaw; // adding yaw correction to integrator correction vector.
// Here we will place a limit on the integrator so that the integrator cannot ever exceed ~30 degrees/second
integrator_magnitude = _omega_I.length();

1
libraries/AP_DCM/AP_DCM.h
View File

@ -126,7 +126,6 @@ private:
Vector3f _omega_I; // Omega Integrator correction
Vector3f _omega_integ_corr; // Partially corrected Gyro_Vector data - used for centrepetal correction
Vector3f _omega; // Corrected Gyro_Vector data
Vector3f _error_roll_pitch;
float _health;
bool _centripetal;
uint8_t _toggle;