dark0707
/
px4-firmware
forked from Archive/PX4-Autopilot
1
0
Fork 0
Code Pull Requests Activity

actual fixes for velocity estimate errors and bad rng fusion

This commit is contained in:
mjturi-mai 2023-02-02 10:13:41 -08:00 committed by Ramon Roche
parent 2dcb525cd9
commit 56c794108d
2 changed files with 25 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
src/modules/ekf2/EKF/optflow_fusion.cpp
View File

@ -77,6 +77,21 @@ void Ekf::updateOptFlow(estimator_aid_source2d_s &aid_src)
const float R_LOS = calcOptFlowMeasVar(_flow_sample_delayed);
aid_src.observation_variance[0] = R_LOS;
aid_src.observation_variance[1] = R_LOS;
const Vector24f state_vector = getStateAtFusionHorizonAsVector();
Vector2f innov_var;
Vector24f H;
sym::ComputeFlowXyInnovVarAndHx(state_vector, P, range, R_LOS, FLT_EPSILON, &innov_var, &H);
innov_var.copyTo(_aid_src_optical_flow.innovation_variance);
// run the innovation consistency check and record result
setEstimatorAidStatusTestRatio(_aid_src_optical_flow, math::max(_params.flow_innov_gate, 1.f));
_innov_check_fail_status.flags.reject_optflow_X = (_aid_src_optical_flow.test_ratio[0] > 1.f);
_innov_check_fail_status.flags.reject_optflow_Y = (_aid_src_optical_flow.test_ratio[1] > 1.f);
// PX4_WARN("updateOptFlow: %f,%f %f,%f\n", (double)aid_src.innovation[0], (double)aid_src.innovation[1], (double)aid_src.innovation_variance[0], (double)aid_src.innovation_variance[1]);
}
void Ekf::fuseOptFlow()

13
src/modules/ekf2/EKF/optical_flow_control.cpp
View File

@ -59,7 +59,8 @@ void Ekf::controlOpticalFlowFusion(const imuSample &imu_delayed)
// Accumulate autopilot gyro data across the same time interval as the flow sensor
const Vector3f delta_angle(imu_delayed.delta_ang - (getGyroBias() * imu_delayed.delta_ang_dt));
if (_delta_time_of < 0.1f) {
// increase limit from 0.1f to 0.5f
if (_delta_time_of < 0.5f) {
_imu_del_ang_of += delta_angle;
_delta_time_of += imu_delayed.delta_ang_dt;
@ -75,7 +76,8 @@ void Ekf::controlOpticalFlowFusion(const imuSample &imu_delayed)
const bool is_tilt_good = (_R_to_earth(2, 2) > _params.range_cos_max_tilt);
const float delta_time_min = fmaxf(0.7f * _delta_time_of, 0.001f);
const float delta_time_max = fminf(1.3f * _delta_time_of, 0.2f);
// increase hard max to 0.5f from 0.2f
const float delta_time_max = fminf(1.3f * _delta_time_of, 0.5f);
bool is_delta_time_good = _flow_sample_delayed.dt >= delta_time_min && _flow_sample_delayed.dt <= delta_time_max;
if (!is_delta_time_good && (_flow_sample_delayed.dt > FLT_EPSILON)) {
@ -204,10 +206,15 @@ void Ekf::controlOpticalFlowFusion(const imuSample &imu_delayed)
if (_time_delayed_us > (_flow_sample_delayed.time_us - uint32_t(1e6f * _flow_sample_delayed.dt) / 2)) {
// Fuse optical flow LOS rate observations into the main filter only if height above ground has been updated recently
// but use a relaxed time criteria to enable it to coast through bad range finder data
if (isRecent(_time_last_hagl_fuse, (uint64_t)10e6)) {
// give much larger timeout (10x) for hagl fuse
if (isRecent(_time_last_hagl_fuse, (uint64_t)100e6)) {
fuseOptFlow();
_last_known_pos.xy() = _state.pos.xy();
}
// else {
// PX4_WARN("terrain estimator sad %ld", _time_last_hagl_fuse);
// }
_flow_data_ready = false;
}