Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Quadrotor having vertical Z up velocity in Altitude hold mode #8109

Closed
siddyyagno3 opened this issue Oct 11, 2017 · 9 comments
Closed

Quadrotor having vertical Z up velocity in Altitude hold mode #8109

siddyyagno3 opened this issue Oct 11, 2017 · 9 comments

Comments

@siddyyagno3
Copy link

siddyyagno3 commented Oct 11, 2017

The Quadrotor has a optical flow px4 flow and Lidar lite V3(i2c). While flying manually the drone is performing normally but when switching to Altitude Hold mode. It starts going up and I have to take manual control. I am using EKF2 in gps denied environment. I also tried changing the aid_mask as well as recalibrating but same.
Should I change to LPE for better output or I am setting some parameter wrong so that distance sensor data is not fusing? Anyway I checked the lidar reading in analyzer topic current distance they are consistent with the actual height. In flight we never went above 4m which is not visible as per local position z. Can this be related to hardware setup?
Please suggest the best setting for a good hover with my current set up.
Flight logs:
https://logs.px4.io/plot_app?log=549f5587-f496-4e81-acb5-ca4b23b629c1

https://logs.px4.io/plot_app?log=1dd97b1d-0ba3-42e2-9e72-05ed7219191d

@AlexisTM
Copy link
Contributor

AlexisTM commented Oct 12, 2017

Hey there:

In both graphs, the Z position is going up and then is reset, while the Z measured distance is correct. Could you try either to disable the barometer or reducing the expected noise of the barometer?

It goes up as the UAV thinks he is falling.

https://logs.px4.io/plot_app?log=1dd97b1d-0ba3-42e2-9e72-05ed7219191d#Nav-Local-Position-Z

https://logs.px4.io/plot_app?log=549f5587-f496-4e81-acb5-ca4b23b629c1#Nav-Local-Position-Z

EKF reports : [lib__ecl] EKF rng hgt timeout - reset to rng hgt

@ChristophTobler & @priseborough

@mad-sci
Copy link

mad-sci commented Oct 12, 2017

Hello @AlexisTM

Below are the parameters related to BARO performance. Which one would you recommend changing, BARO_NOISE or BETA_NOISE be increased or decreased? Or is there any other parameter which also needs modification.

EKF2_BARO_DELAY, 0.0
EKF2_BARO_GATE, 5.0
EKF2_BARO_NOISE, 5.0
EKF2_BETA_NOISE, 0.30000001192092896

@ChristophTobler
Copy link
Contributor

ChristophTobler commented Oct 13, 2017

@siddyyagno3 It looks like your Hardware is not dampened correctly. Can you send a picture of your hardware? Have a look at your acceleration (should look more like this). The ekf2's height estimation is quite sensible to noisy IMU data. Can you try fixing your dampening and check if it still happens?

@ChristophTobler
Copy link
Contributor

ChristophTobler commented Oct 13, 2017

This has nothing to do with barometer or barometer parameters as EKF2_HGT_MODE was set to range finder (2). Therefore it uses the distance sensor for height estimation and not the barometer.

@siddyyagno3
Copy link
Author

siddyyagno3 commented Oct 13, 2017

@ChristophTobler
It looks like this may be the issue but I also had that intuition that is why I also tested a previous flight with EKF2_aid_mask set to inhibit IMU bias but the result was same. Doesn't this parameter settings takes IMU acceleration out of the picture?
I don't have flight data to support that. I will redo the same flight. Subsequently,
@mad-sci we will improve the damping and lower the vibration and upload the test flight data as the raw acceleration do look very bad in our case.

@ChristophTobler
Copy link
Contributor

@siddyyagno3 Unfortunately there is no parameter that magically solves mechanical noise/issues ;)
Did you ever do a "normal" flight outdoors using just GPS and barometer? Would be surprised if that works.

@priseborough
Copy link
Contributor

I agree with @ChristophTobler. Your vibration levels are much too high. You need to isolate your autopilot board from the airframe using soft mounts and make sure your connecting cables have adequate slack and bend in them so they also do not transmit vibration.

This is not a firmware problem.

@mad-sci
Copy link

mad-sci commented Oct 13, 2017

@ChristophTobler @AlexisTM @priseborough Thank you for your inputs and got better results after changing the dampers of the Pixhawk. Still the local position Z (https://logs.px4.io/plot_app?log=ebf3160c-8049-4971-b362-f761db190f2e#Nav-Local-Position-Z)is not good enough as in this (https://logs.px4.io/plot_app?log=88e4ba1c-8f73-4b66-b988-c7cbd34f7739#Nav-Local-Position-Z)
Do we need to change any other parameters?

My Flight logs https://logs.px4.io/plot_app?log=ebf3160c-8049-4971-b362-f761db190f2e

@ranjeetkumar16722
Copy link

In altitude hold mode, vehicle will ascend to the height specified by parameter MIS_LTRMIN_ALT and hold at that altitude. Please check the parameter what you have set.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants