floam icon indicating copy to clipboard operation
floam copied to clipboard

Ouster lidar 64 support

Open islamtalha01 opened this issue 2 years ago • 9 comments

Hi, Can you tell me that which line of code I needed to use this package for the ouster os1 64? I need to use this package with a rosbag recorded with point cloud and image topic. I need to use it for the automatic calibration package.i want to get dense point cloud in my rosbag so is the first setup is that to setup my sensor again and launch driver using floam launch file and then record again or I can just use rosbag file for this purpose?

islamtalha01 avatar Oct 13 '21 05:10 islamtalha01

Hi, I have a similar request: I would like to use FLAOM with a Ouster OS0-64 dataset. I can see there are some hard coded angles here: https://github.com/wh200720041/floam/blob/9fdb2ca60a2988c0db515c9c1567100fba8b58d8/src/laserProcessingClass.cpp#L48-L59 which seem to match Kitti dataset Velodyne HDL-64E sensor (datasheet) But I am unsure which values I have to change. The Ouster OS0-64 also has 64 scan lines, but a vertical field of view of -45° .. +45°. Your help is appreciated.

masterhui avatar Oct 15 '21 13:10 masterhui

hi @masterhui and @islamtalha01

from ouster website, note that there are several configuration as shown below

Screenshot 2021-10-16 17:54:23

By default, it should be uniform beam spacing. We use uniform beam for example. else if (N_SCANS == 64) {
scanID = round((angle + 22.5) * 45.0 / 63.0); if (scanID > (N_SCANS - 1) || scanID < 0) continue;

}

However, for those who use OS0 and OS2, The harware configuration is different, OS2 is 22.5 and OS0 is 90, please change accordingly based on the code above.

Hence, for OS0, it is else if (N_SCANS == 64) {
scanID = round((angle + 45) * 90.0 / 63.0); if (scanID > (N_SCANS - 1) || scanID < 0) continue; }

You can leave you question here for different sensor support.

wh200720041 avatar Oct 16 '21 10:10 wh200720041

Hello @wh200720041 thank you for your tips. I have followed your advice and replaced the above lines in laserProcessingClass.cpp for use with my Ouster OS0-64, which is configured for uniform 90° vertical field of view. However, after re-compiling the results look much worse than when using the original Velodyne code.

Map with original code: Screenshot from 2021-10-18 12-52-55

Map with Ouster OS0-64 code: Screenshot from 2021-10-18 12-41-02

Any ideas why this might be the case? Regards

masterhui avatar Oct 18 '21 11:10 masterhui

which IMU did you use? The built-in IMU?

juliangaal avatar Jan 20 '22 11:01 juliangaal

Thank you for your response. i have closed that topic. I think it was built in IMU but not sure

On Thu, Jan 20, 2022 at 4:20 PM Julian Gaal @.***> wrote:

which IMU did you use? The built-in IMU?

— Reply to this email directly, view it on GitHub https://github.com/wh200720041/floam/issues/46#issuecomment-1017396974, or unsubscribe https://github.com/notifications/unsubscribe-auth/AT4S27GPXY7BZLRZLDKQW6DUW7VYXANCNFSM5F4IH7EQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

islamtalha01 avatar Jan 20 '22 19:01 islamtalha01

Hi, here is a quick update. I managed to get this to work perfectly with data from my Ouster OS0-64. In case someone else is looking for a solution, please see patch below. Settings-for-Ouster-OS0-64-90.patch.txt

masterhui avatar Jan 21 '22 14:01 masterhui

Thanks, I'll check it out.

On Fri, Jan 21, 2022 at 7:43 PM masterhui @.***> wrote:

Hi, here is a quick update. I managed to get this to work perfectly with data from my Ouster OS0-64. In case someone else is looking for a solution, please see patch below. Settings-for-Ouster-OS0-64-90.patch.txt https://github.com/wh200720041/floam/files/7914461/Settings-for-Ouster-OS0-64-90.patch.txt

— Reply to this email directly, view it on GitHub https://github.com/wh200720041/floam/issues/46#issuecomment-1018567969, or unsubscribe https://github.com/notifications/unsubscribe-auth/AT4S27CKMKQMIAWQIXETM7DUXFWHTANCNFSM5F4IH7EQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

islamtalha01 avatar Jan 21 '22 16:01 islamtalha01

Hi, here is a quick update. I managed to get this to work perfectly with data from my Ouster OS0-64. In case someone else is looking for a solution, please see patch below. Settings-for-Ouster-OS0-64-90.patch.txt

Are you using the built-in IMU of the Ouster?

juliangaal avatar Jan 21 '22 23:01 juliangaal

Not sure if I understand your question. One of the features of F-LOAM is that it doesn't require an IMU, am I wrong?

masterhui avatar Feb 01 '22 14:02 masterhui

I have a gradient OS1-64, what configuration should I use?

gglaspell avatar Jun 09 '22 17:06 gglaspell

Hi @gglaspell Check out the patch I posted above for the OS0-64. You will have to modify the values as the OS1 has different vertical opening angles.

masterhui avatar Jun 20 '22 09:06 masterhui