New filter for distance sensor validation #24106

haumarco · 2024-12-13T13:22:30Z

Solved Problem

The distance sensor is being fused or used to reset the terrain state, even though its measurements do not accurately represent the distance to the ground due to obstructions caused by rain, snow, fog etc.

Solution

Now the distance sensor can be in 3 different states based on the lpf test ratio of a simple kalman filter: consistent, inconsistent and unknown. A terrain reset can only happen when the state is consistent. The sensor will be used for normal fusion in both consistent and inconsistent states. Together with the already existing fog detector, the robustness against sensor blockage/failure is increased.

new:

old:

github-actions · 2024-12-13T13:28:06Z

🔎 FLASH Analysis

px4_fmu-v5x [Total VM Diff: 1016 byte (0.05 %)]

    FILE SIZE        VM SIZE    
--------------  -------------- 
+0.1%   +1016  +0.1%   +1016    .text
   +21%    +440   +21%    +440    Ekf::controlRangeHaglFusion()
  +131%    +388  +131%    +388    RangeFinderConsistencyCheck::update()
  [NEW]    +116  [NEW]    +116    RangeFinderConsistencyCheck::evaluateState()
  [NEW]     +64  [NEW]     +64    RangeFinderConsistencyCheck::init()
  +1.4%     +48  +1.4%     +48    Ekf::Ekf()
  +0.0%     +20  +0.0%     +20    g_cromfs_image
  +1.5%     +20  +1.5%     +20    ucdr_serialize_estimator_status_flags()
  +4.8%     +16  +4.8%     +16    EKF2::PublishInnovationTestRatios()
  +4.7%     +16  +4.7%     +16    EKF2::PublishInnovationVariances()
  +4.8%     +16  +4.8%     +16    EKF2::PublishInnovations()
 -99.4%     +16 -99.4%     +16    [6 Others]
  +0.8%      +8  +0.8%      +8    EKF2::PublishStatusFlags()
  +2.1%      +8  +2.1%      +8    Ekf::controlFakeHgtFusion()
  +0.1%      +8  +0.1%      +8    Ekf::predictCovariance()
  +0.0%      +8  +0.0%      +8    [section .text]
  +0.0%      +8  +0.0%      +8    uORB::compressed_fields
  +0.3%      +4  +0.3%      +4    EKF2::AdvertiseTopics()
  -0.2%     -12  -0.2%     -12    Ekf::controlMagFusion()
  -5.5%     -16  -5.5%     -16    Ekf::updateRangeHagl()
  -0.9%     -20  -0.9%     -20    Ekf::fuseMag()
  [DEL]    -140  [DEL]    -140    RangeFinderConsistencyCheck::updateConsistency()
+0.0%     +82  [ = ]       0    .debug_abbrev
+0.0%     +16  [ = ]       0    .debug_aranges
+0.0%    +172  [ = ]       0    .debug_frame
+0.0% +10.0Ki  [ = ]       0    .debug_info
+0.1% +5.42Ki  [ = ]       0    .debug_line
   +33%      +1  [ = ]       0    [Unmapped]
  +0.1% +5.42Ki  [ = ]       0    [section .debug_line]
+0.2% +5.61Ki  [ = ]       0    .debug_loclists
+0.2% +1.30Ki  [ = ]       0    .debug_rnglists
  [DEL]      -1  [ = ]       0    [Unmapped]
  +0.2% +1.31Ki  [ = ]       0    [section .debug_rnglists]
+0.0%    +412  [ = ]       0    .debug_str
-0.4%      -1  [ = ]       0    .shstrtab
+0.0%     +41  [ = ]       0    .strtab
  [NEW]     +52  [ = ]       0    RangeFinderConsistencyCheck::evaluateState()
  [NEW]     +43  [ = ]       0    RangeFinderConsistencyCheck::init()
  +2.1%      +1  [ = ]       0    RangeFinderConsistencyCheck::update()
  [DEL]     -55  [ = ]       0    RangeFinderConsistencyCheck::updateConsistency()
 -43.2%     -16  [ = ]       0    __sq_addafter_veneer
   +67%     +16  [ = ]       0    __uart_connected_veneer
+0.0%     +48  [ = ]       0    .symtab
   +33%     +16  [ = ]       0    EKF2::PublishEventFlags()
  +200%     +32  [ = ]       0    EKF2::PublishInnovationTestRatios()
  +100%     +16  [ = ]       0    EKF2::PublishInnovations()
 -50.0%     -16  [ = ]       0    EKF2::updateParamsImpl()
   +20%     +16  [ = ]       0    Ekf::controlRangeHaglFusion()
 -25.0%     -16  [ = ]       0    Ekf::isYawFailure()
 -50.0%     -16  [ = ]       0    Ekf::resetMagBiasCov()
   +50%     +16  [ = ]       0    Ekf::resetMagEarthCov()
  +100%     +16  [ = ]       0    Ekf::stopMagFusion()
 -66.7%     -32  [ = ]       0    Ekf::updateRangeHagl()
   +33%     +16  [ = ]       0    EstimatorInterface::setSystemFlagData()
  [NEW]     +48  [ = ]       0    RangeFinderConsistencyCheck::evaluateState()
  [NEW]     +48  [ = ]       0    RangeFinderConsistencyCheck::init()
   +33%     +16  [ = ]       0    RangeFinderConsistencyCheck::update()
  [DEL]     -48  [ = ]       0    RangeFinderConsistencyCheck::updateConsistency()
 -33.3%     -32  [ = ]       0    RcvTopicsPubs::init()
 -50.0%     +64  [ = ]       0    [3 Others]
  -0.3%     -32  [ = ]       0    [section .symtab]
 -40.0%     -32  [ = ]       0    __sq_addafter_veneer
 -25.0%     -16  [ = ]       0    __stm32_freesegment_veneer
 -33.3%     -16  [ = ]       0    __stm32_i2c_disable_reload_veneer
-10.3%   -1016  [ = ]       0    [Unmapped]
+0.1% +23.1Ki  +0.0%   +1016    TOTAL

px4_fmu-v6x [Total VM Diff: 984 byte (0.05 %)]

    FILE SIZE        VM SIZE    
--------------  -------------- 
+0.1%    +984  +0.1%    +984    .text
   +21%    +440   +21%    +440    Ekf::controlRangeHaglFusion()
  +131%    +388  +131%    +388    RangeFinderConsistencyCheck::update()
  [NEW]    +116  [NEW]    +116    RangeFinderConsistencyCheck::evaluateState()
  [NEW]     +64  [NEW]     +64    RangeFinderConsistencyCheck::init()
  +1.4%     +48  +1.4%     +48    Ekf::Ekf()
  +1.5%     +20  +1.5%     +20    ucdr_serialize_estimator_status_flags()
  +4.8%     +16  +4.8%     +16    EKF2::PublishInnovationTestRatios()
  +4.7%     +16  +4.7%     +16    EKF2::PublishInnovationVariances()
  +4.8%     +16  +4.8%     +16    EKF2::PublishInnovations()
 -100.0%     +12 -100.0%     +12    [5 Others]
  +0.8%      +8  +0.8%      +8    EKF2::PublishStatusFlags()
  +2.1%      +8  +2.1%      +8    Ekf::controlFakeHgtFusion()
  +0.1%      +8  +0.1%      +8    Ekf::predictCovariance()
  +0.0%      +8  +0.0%      +8    uORB::compressed_fields
  +0.3%      +4  +0.3%      +4    EKF2::AdvertiseTopics()
  +4.5%      +4  +4.5%      +4    Ekf::resetMagBiasCov()
  -0.7%      -4  -0.7%      -4    Ekf::controlFusionModes()
  -0.2%     -12  -0.2%     -12    Ekf::controlMagFusion()
  -5.5%     -16  -5.5%     -16    Ekf::updateRangeHagl()
  -0.9%     -20  -0.9%     -20    Ekf::fuseMag()
  [DEL]    -140  [DEL]    -140    RangeFinderConsistencyCheck::updateConsistency()
+0.0%     +82  [ = ]       0    .debug_abbrev
+0.0%     +16  [ = ]       0    .debug_aranges
+0.0%    +172  [ = ]       0    .debug_frame
+0.0% +10.0Ki  [ = ]       0    .debug_info
+0.1% +5.42Ki  [ = ]       0    .debug_line
   +50%      +1  [ = ]       0    [Unmapped]
  +0.1% +5.42Ki  [ = ]       0    [section .debug_line]
+0.2% +5.60Ki  [ = ]       0    .debug_loclists
+0.2% +1.30Ki  [ = ]       0    .debug_rnglists
 -66.7%      -2  [ = ]       0    [Unmapped]
  +0.2% +1.31Ki  [ = ]       0    [section .debug_rnglists]
+0.0%    +412  [ = ]       0    .debug_str
+1.3%      +3  [ = ]       0    .shstrtab
+0.0%     +41  [ = ]       0    .strtab
  [NEW]     +52  [ = ]       0    RangeFinderConsistencyCheck::evaluateState()
  [NEW]     +43  [ = ]       0    RangeFinderConsistencyCheck::init()
  +2.1%      +1  [ = ]       0    RangeFinderConsistencyCheck::update()
  [DEL]     -55  [ = ]       0    RangeFinderConsistencyCheck::updateConsistency()
+0.0%     +48  [ = ]       0    .symtab
   +33%     +16  [ = ]       0    EKF2::PublishEventFlags()
  +200%     +32  [ = ]       0    EKF2::PublishInnovationTestRatios()
  +100%     +16  [ = ]       0    EKF2::PublishInnovations()
 -50.0%     -16  [ = ]       0    EKF2::updateParamsImpl()
   +20%     +16  [ = ]       0    Ekf::controlRangeHaglFusion()
 -25.0%     -16  [ = ]       0    Ekf::isYawFailure()
 -50.0%     -16  [ = ]       0    Ekf::resetMagBiasCov()
   +50%     +16  [ = ]       0    Ekf::resetMagEarthCov()
  +100%     +16  [ = ]       0    Ekf::stopMagFusion()
 -66.7%     -32  [ = ]       0    Ekf::updateRangeHagl()
   +33%     +16  [ = ]       0    EstimatorInterface::setSystemFlagData()
  [NEW]     +48  [ = ]       0    RangeFinderConsistencyCheck::evaluateState()
  [NEW]     +48  [ = ]       0    RangeFinderConsistencyCheck::init()
   +33%     +16  [ = ]       0    RangeFinderConsistencyCheck::update()
  [DEL]     -48  [ = ]       0    RangeFinderConsistencyCheck::updateConsistency()
 -33.3%     -32  [ = ]       0    RcvTopicsPubs::init()
  -0.3%     -32  [ = ]       0    [section .symtab]
-12.6%    -984  [ = ]       0    [Unmapped]
+0.1% +23.1Ki  +0.0%    +984    TOTAL

Updated: 2025-05-19T13:01:29

bresch

I'm wondering if using sequential fusion would actually give nicer code in the end so you wouldn't need to build all those 2x2 matrices

src/modules/ekf2/EKF/python/ekf_derivation/derivation.py

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.hpp

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

AlexKlimaj · 2024-12-27T02:11:28Z

@dakejahl Does this help our height issues?

AlexKlimaj · 2024-12-31T21:06:31Z

This was a quick indoor flight in manual mode.

https://review.px4.io/plot_app?log=e3d3cf76-563d-4a23-9308-d37cbdcd34c3

Vs current main

https://review.px4.io/plot_app?log=6d902213-02ef-4ff8-8474-d341bfcc7bd9

haumarco · 2025-01-10T09:50:30Z

@AlexKlimaj Your vertical velocity estimate seems to be off quite a lot. Can you increase the bias noise on the acc so the bias will converge a bit quicker? The new filter uses the velocity estimate of the EKF to verify if the measurement is correct or not but when the EKF estimate is off and has a low uncertainty then it will discard the range measurements.

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

src/modules/ekf2/EKF/python/ekf_derivation/derivation.py

src/modules/ekf2/EKF/aid_sources/range_finder/range_height_control.cpp

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

src/modules/ekf2/EKF/python/ekf_derivation/derivation.py

src/modules/ekf2/test/test_EKF_height_fusion.cpp

src/modules/ekf2/test/test_EKF_terrain.cpp

bresch · 2025-01-28T15:46:02Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_height_control.cpp

@@ -268,11 +278,9 @@ void Ekf::updateRangeHagl(estimator_aid_source1d_s &aid_src)

 float Ekf::getRngVar() const
 {
-	return fmaxf(
-		       P(State::pos.idx + 2, State::pos.idx + 2)


src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.hpp

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

src/modules/ekf2/EKF/covariance.cpp

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

bresch · 2025-02-10T08:50:07Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

-			_time_last_inconsistent_us = time_us;
+	// start the consistency check after 1s
+	if (_t_since_first_sample + dt > 1.0f) {
+		_t_since_first_sample = 2.0f;


Do you really need to force it to 2.0? Maybe remove the +dt in the condition so when it enters the if once, it will always enter it (even if dt is suddenly slightly lower).

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

dakejahl · 2025-04-05T04:46:46Z

Did a quick flight indoors with the Dexi (Flow, range, and baro. EKF2_HGT_REF=1). Lots of wandering up and down, have to fight with the throttle to keep the altitude in place
https://review.px4.io/plot_app?log=e0095444-5705-490b-82e3-ea1efb4784e5

haumarco · 2025-04-22T11:51:02Z

@dakejahl if i replay your log with main it looks pretty much the same. seems like the classic indoor z/vz inconsistency based on suboptimal bias-uncertainties. Or do you have a log where this is not the issue?

dakejahl · 2025-04-23T02:13:33Z

Okay let's take a step back and just focus on outdoor flights.

@haumarco Here's an outdoor flight. EKF replay is enabled in this log. This is based on main from today with this PR and #24751 merged into it.

Dexi airframe (quad)

No GPS
Distance sensor
Optical Flow
Barometer

Mostly default params except reduced EKF2_RNG_NOISE to 0.01. So the EKF2_HGT_REF is gps (so it fallbacks to baro), although throughout most of the flight "range aiding" was active, but doesn't appear to work very well since the altitude is fluctuating quite a bit (there was a little bit of wind). I kept the throttle around 0 for most of the flight, for simply testing altitude hold performance. MPC_HOLD_DZ is left at the default of 0.1.

https://review.px4.io/plot_app?log=ddbf2f12-14a9-4400-815c-2520cd347313

DronecodeBot · 2025-04-23T03:08:51Z

This pull request has been mentioned on Discussion Forum for PX4, Pixhawk, QGroundControl, MAVSDK, MAVLink. There might be relevant details there:

https://discuss.px4.io/t/px4-sync-q-a-april-23-2025/45198/3

dakejahl · 2025-05-05T00:29:22Z

We should get this in. I've got a branch that improves altitude hold with a rangefinder which is based off of this work.

dakejahl · 2025-05-06T05:58:41Z

@haumarco can you share the log those screenshots in the description are from? Or any other logs you have of this PR with a distance sensor would be helpful

* move P-init to ekf-derivation script

…. this fixes behavior for jumps in fixed wing

…n is not defined in test

bresch · 2025-05-26T14:04:11Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.hpp

+	KinematicState _state{KinematicState::UNKNOWN};
+	float _t_since_first_sample{0.f};
+	uint8_t _last_posD_reset_count{0};
+	static constexpr float _t_to_init{1.f};


Suggested change

static constexpr float _t_to_init{1.f};

static constexpr float _kTestRatioLpfTimeConstant{1.f};

bresch · 2025-05-26T14:04:55Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

+		} else if (measurement_idx == 1) {
+			_innov = y;
+			const float test_ratio = fminf(sq(y) / (sq(_gate) * S), 4.f); // limit to 4 to limit sensitivity to outliers
+			_test_ratio_lpf.setParameters(dt, _t_to_init);


you can set the time constant when declaring the AlphaFilter and then call the update(dt) function

bresch · 2025-05-26T14:09:07Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp

-		if ((time_us - _time_last_horizontal_motion) > _signed_test_ratio_tau) {
-			_is_kinematically_consistent = false;
-			_time_last_inconsistent_us = time_us;
+	// start the consistency check after 1s


Suggested change

// start the consistency check after 1s

// let the filter settle before starting the consistency check

Avoid mentioning numerical values defined elsewhere as ppl will change them without updating the comments

bresch · 2025-05-26T14:19:06Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.hpp

+	uint8_t current_posD_reset_count{0};
+	bool horizontal_motion{false};


Except if not possible otherwise, member variables should always be private and set/access with setter and getter functions. (and prefixed with _)

bresch · 2025-05-26T14:21:35Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.hpp

@@ -48,36 +48,62 @@ class RangeFinderConsistencyCheck final
 	RangeFinderConsistencyCheck() = default;
 	~RangeFinderConsistencyCheck() = default;

-	void update(float dist_bottom, float dist_bottom_var, float vz, float vz_var, bool horizontal_motion, uint64_t time_us);
+	enum class KinematicState : int {


https://google.github.io/styleguide/cppguide.html#Enumerator_Names

bresch · 2025-05-26T14:35:37Z

src/modules/ekf2/EKF/aid_sources/range_finder/range_height_control.cpp


+				if (!updated_horizontal_motion && _rng_consistency_check.horizontal_motion) {


the consistency check class itself can trigger the reset when the motion->no_motion transition is detected ("tell don't ask principle": https://martinfowler.com/bliki/TellDontAsk.html)

bresch · 2025-05-26T15:01:43Z

src/modules/ekf2/test/test_EKF_accelerometer.cpp


 	// AND: an accelerometer with a really large Z bias
 	const float bias = CONSTANTS_ONE_G;
 	_sensor_simulator._imu.setAccelData(Vector3f(0.f, 0.f, -CONSTANTS_ONE_G + bias));
 	_sensor_simulator.runSeconds(2);
+	_sensor_simulator._imu.setAccelClipping(false, false, true);


Wait, the unit test is to make sure the ekf detect when there is an IMU problem using baro and the range finder, you just changed it to detect the accel clipping flag, this is a different check.

bresch · 2025-05-26T15:03:31Z

src/modules/ekf2/test/test_EKF_height_fusion.cpp

@@ -162,7 +155,7 @@ TEST_F(EkfHeightFusionTest, gpsRef)
 	_ekf_wrapper.enableBaroHeightFusion();
 	_ekf_wrapper.enableGpsHeightFusion();
 	_ekf_wrapper.enableRangeHeightFusion();
-	_sensor_simulator.runSeconds(1);
+	_sensor_simulator.runSeconds(10);


Does it just take more time to start the range height fusion now?

bresch · 2025-05-26T15:03:57Z

src/modules/ekf2/test/test_EKF_height_fusion.cpp

@@ -242,7 +235,7 @@ TEST_F(EkfHeightFusionTest, baroRefFailOver)
 	_ekf_wrapper.enableGpsHeightFusion();
 	_ekf_wrapper.enableRangeHeightFusion();

-	_sensor_simulator.runSeconds(0.1);
+	_sensor_simulator.runSeconds(10);


100 times slower?

bresch · 2025-05-26T15:05:13Z

src/modules/ekf2/test/test_EKF_terrain.cpp

@@ -207,7 +207,7 @@ TEST_F(EkfTerrainTest, testHeightReset)
 	const float new_baro_height = _sensor_simulator._baro.getData() + 50.f;
 	_sensor_simulator._baro.setData(new_baro_height);
 	_sensor_simulator.stopGps(); // prevent from switching to GNSS height
-	_sensor_simulator.runSeconds(10);
+	_sensor_simulator.runSeconds(100);


Do you really need 100s? Isn't something broken here?

haumarco requested a review from bresch December 13, 2024 13:22

bresch requested changes Dec 13, 2024

View reviewed changes

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch 3 times, most recently from 499efdc to f7052d6 Compare December 20, 2024 10:16

haumarco marked this pull request as ready for review December 20, 2024 12:06

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch 2 times, most recently from 11655eb to 9f42b6e Compare December 20, 2024 12:58

bresch reviewed Dec 20, 2024

View reviewed changes

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp Outdated Show resolved Hide resolved

bresch reviewed Dec 20, 2024

View reviewed changes

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp Outdated Show resolved Hide resolved

bresch reviewed Dec 20, 2024

View reviewed changes

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp Outdated Show resolved Hide resolved

bresch reviewed Dec 20, 2024

View reviewed changes

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp Outdated Show resolved Hide resolved

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch from 9f42b6e to 42ea74c Compare December 30, 2024 09:20

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch from 42ea74c to e488ddb Compare January 10, 2025 09:55

dagar added the EKF2 label Jan 15, 2025

bresch requested changes Jan 24, 2025

View reviewed changes

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch 2 times, most recently from b0361da to d07902d Compare January 29, 2025 08:32

bresch requested changes Jan 29, 2025

View reviewed changes

bresch reviewed Jan 29, 2025

View reviewed changes

src/modules/ekf2/EKF/aid_sources/range_finder/range_finder_consistency_check.cpp Outdated Show resolved Hide resolved

bresch requested changes Feb 3, 2025

View reviewed changes

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch from 5e19f1b to 005510f Compare February 4, 2025 07:59

bresch requested changes Feb 10, 2025

View reviewed changes

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch 2 times, most recently from ae44720 to 8946cd2 Compare March 10, 2025 08:47

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch from 8946cd2 to 95053b6 Compare April 4, 2025 18:01

dakejahl force-pushed the pr-ekf2_dist_sensor_validation branch from 95053b6 to 5659535 Compare May 7, 2025 00:50

bresch self-requested a review May 19, 2025 08:17

haumarco added 19 commits May 19, 2025 14:45

add mini KF to validate range sensor measurements

47b163c

handle noise while hovering

5a9da7d

fix wrong miniEKF init

bd746d7

fix bugs, rejection was too sensitive

16eeddc

always update lpf

43ed8cb

add symforce, adjust smaller things

2964f4d

adjust threshold unit tests

abb0b4c

overhaul approach to work on unit tests and snow/blocking/normal flights

0fa4cd4

reset state to unknown on init

c8e43db

clean up range-filter fusion step

f81b88e

fix uncertainty initialization

512e555

* clean up method calls

3d14318

* move P-init to ekf-derivation script

change consistency to only be valid if running for a second

adad028

also run range consistency check while landed

dda3f24

adjust IMU falling detection baro-range unittest

4c0e03f

only run range validator on regular data

10adbb3

minor cleanup

a872c30

always run consistency-KF but limit state change to vertical velocity…

bd8e457

…. this fixes behavior for jumps in fixed wing

remove terrain checks form baroRef unit-tests, vehicle vertical motio…

88cb887

…n is not defined in test

haumarco force-pushed the pr-ekf2_dist_sensor_validation branch from 5659535 to 88cb887 Compare May 19, 2025 12:55

bresch requested changes May 26, 2025

View reviewed changes

dakejahl mentioned this pull request May 28, 2025

[Bug] Unstable height estimates #24058

Open

	static constexpr float _t_to_init{1.f};
	static constexpr float _kTestRatioLpfTimeConstant{1.f};

	// start the consistency check after 1s
	// let the filter settle before starting the consistency check

		uint8_t current_posD_reset_count{0};
		bool horizontal_motion{false};


		if (!updated_horizontal_motion && _rng_consistency_check.horizontal_motion) {

New filter for distance sensor validation #24106

Are you sure you want to change the base?

New filter for distance sensor validation #24106

Conversation

haumarco commented Dec 13, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Solved Problem

Solution

Uh oh!

github-actions bot commented Dec 13, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔎 FLASH Analysis

Uh oh!

bresch left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

AlexKlimaj commented Dec 27, 2024

Uh oh!

AlexKlimaj commented Dec 31, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

haumarco commented Jan 10, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

dakejahl commented Apr 5, 2025

Uh oh!

haumarco commented Apr 22, 2025

Uh oh!

dakejahl commented Apr 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

DronecodeBot commented Apr 23, 2025

Uh oh!

dakejahl commented May 5, 2025

Uh oh!

dakejahl commented May 6, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

haumarco commented Dec 13, 2024 •

edited

Loading

github-actions bot commented Dec 13, 2024 •

edited

Loading

AlexKlimaj commented Dec 31, 2024 •

edited

Loading

dakejahl commented Apr 23, 2025 •

edited

Loading