How is data synchronization achieved between the lidar and the GNSS module?

[JACKLiuDay]

Dr. He, could you explain how lidar and GNSS data synchronization is implemented?I currently have a Mid-360 radar and a ZED-F9P GNSS module—how can I achieve data synchronization between them?

[Joanna-HE]

We synchronize the GNSS with LiDAR through PPS, as stated in the github repository. The LiDAR used in our self-collected datasets supports receiving PPS signals and GPRMC data, which can be published through serial ports by the GNSS receivers, like u-blox F9P and others. After receiving the PPS signals and GPRMC data, the LiDAR will adjust the package and time of the package to be synchronized with the time included in the GPRMC data.

…

________________________________ 寄件者: Dayu ***@***.***> 寄件日期: 2025年4月13日 下午 03:39 收件者: Joanna-HE/LIGO. ***@***.***> 副本: Subscribed ***@***.***> 主旨: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) This is an external email. Dr. He, could you explain how lidar and GNSS data synchronization is implemented? — Reply to this email directly, view it on GitHub<#17>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXXISNNMX2AZR7QIPY32ZIICZAVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43ASLTON2WKOZSHE4TCMJQGY2DAOI>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> [https://avatars.githubusercontent.com/u/30967867?s=20&v=4]JACKLiuDay created an issue (Joanna-HE/LIGO.#17)<#17> Dr. He, could you explain how lidar and GNSS data synchronization is implemented? — Reply to this email directly, view it on GitHub<#17>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXXISNNMX2AZR7QIPY32ZIICZAVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43ASLTON2WKOZSHE4TCMJQGY2DAOI>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[JACKLiuDay]

We synchronize the GNSS with LiDAR through PPS, as stated in the github repository. The LiDAR used in our self-collected datasets supports receiving PPS signals and GPRMC data, which can be published through serial ports by the GNSS receivers, like u-blox F9P and others. After receiving the PPS signals and GPRMC data, the LiDAR will adjust the package and time of the package to be synchronized with the time included in the GPRMC data.
…

Dr. He, I checked the radar frames and the range_meas topic in the bridge.bag file, and their timestamps don't seem to match. Does this mean the GNSS and LiDAR data in this bag file have already been timestamp-synchronized, or was no hardware time synchronization performed?

Additionally, in the avia.yaml file, gnss_local_online_sync is set to false, and gnss_local_time_diff is set to 18.0. What is the meaning of these settings?

[Joanna-HE]

I cannot understand “their timestamps don't seem to match". We definitely apply the time synchronization. The setting is for ligo system, the ligo doesn't use the online updated synchronized time for running, since 18s is accurate enough. This time difference, i.e., 18s, between gnss message and the LiDAR indicates that we synchronized those two sensors. 获取 Outlook for iOS<https://aka.ms/o0ukef>

…

________________________________ 发件人: Dayu ***@***.***> 发送时间: Friday, April 18, 2025 10:20:12 AM 收件人: Joanna-HE/LIGO. ***@***.***> 抄送: ***@***.*** ***@***.***>; Comment ***@***.***> 主题: Re: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) This is an external email. We synchronize the GNSS with LiDAR through PPS, as stated in the github repository. The LiDAR used in our self-collected datasets supports receiving PPS signals and GPRMC data, which can be published through serial ports by the GNSS receivers, like u-blox F9P and others. After receiving the PPS signals and GPRMC data, the LiDAR will adjust the package and time of the package to be synchronized with the time included in the GPRMC data. … image.png (view on web)<https://github.com/user-attachments/assets/a8030a8e-16e9-4f28-b059-0b2589081c32> Dr. He, I checked the radar frames and the range_meas topic in the bridge.bag file, and their timestamps don't seem to match. Does this mean the GNSS and LiDAR data in this bag file have already been timestamp-synchronized, or was no hardware time synchronization performed? Additionally, in the avia.yaml file, gnss_local_online_sync is set to false, and gnss_local_time_diff is set to 18.0. What is the meaning of these settings? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXVTTE5NZQ7SIU7ISU322BONZAVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJUGM3DQMBRHA>. You are receiving this because you commented.Message ID: ***@***.***> [https://avatars.githubusercontent.com/u/30967867?s=20&v=4]JACKLiuDay left a comment (Joanna-HE/LIGO.#17)<#17 (comment)> We synchronize the GNSS with LiDAR through PPS, as stated in the github repository. The LiDAR used in our self-collected datasets supports receiving PPS signals and GPRMC data, which can be published through serial ports by the GNSS receivers, like u-blox F9P and others. After receiving the PPS signals and GPRMC data, the LiDAR will adjust the package and time of the package to be synchronized with the time included in the GPRMC data. … image.png (view on web)<https://github.com/user-attachments/assets/a8030a8e-16e9-4f28-b059-0b2589081c32> Dr. He, I checked the radar frames and the range_meas topic in the bridge.bag file, and their timestamps don't seem to match. Does this mean the GNSS and LiDAR data in this bag file have already been timestamp-synchronized, or was no hardware time synchronization performed? Additionally, in the avia.yaml file, gnss_local_online_sync is set to false, and gnss_local_time_diff is set to 18.0. What is the meaning of these settings? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXVTTE5NZQ7SIU7ISU322BONZAVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJUGM3DQMBRHA>. You are receiving this because you commented.Message ID: ***@***.***>

[JACKLiuDay]

I cannot understand “their timestamps don't seem to match". We definitely apply the time synchronization.

The setting is for ligo system, the ligo doesn't use the online updated synchronized time for running, since 18s is accurate enough. This time difference, i.e., 18s, between gnss message and the LiDAR indicates that we synchronized those two sensors.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Sorry, Dr. He, I didn’t quite understand your point. If the GNSS data and LiDAR have already been hardware-synchronized, then why is there still an 18-second delay that needs to be set?

[Joanna-HE]

Yes, this is the time difference between the gps time that included in the gnss message, and the utc time that is included in the gprmc message. The gprmc is sent to LiDAR for lidar time synchronization, giving lidar utc time, while gnss message uses gps time. 获取 Outlook for iOS<https://aka.ms/o0ukef>

…

________________________________ 发件人: Dayu ***@***.***> 发送时间: Monday, April 21, 2025 11:29:03 AM 收件人: Joanna-HE/LIGO. ***@***.***> 抄送: ***@***.*** ***@***.***>; Comment ***@***.***> 主题: Re: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) This is an external email. I cannot understand “their timestamps don't seem to match". We definitely apply the time synchronization. The setting is for ligo system, the ligo doesn't use the online updated synchronized time for running, since 18s is accurate enough. This time difference, i.e., 18s, between gnss message and the LiDAR indicates that we synchronized those two sensors. 获取 Outlook for iOShttps://aka.ms/o0ukef … Sorry, Dr. He, I didn’t quite understand your point. If the GNSS data and LiDAR have already been hardware-synchronized, then why is there still an 18-second delay that needs to be set? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXUNVS3P7U4RD74PXVL22RQX7AVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJXGU2TIOJSGI>. You are receiving this because you commented.Message ID: ***@***.***> [https://avatars.githubusercontent.com/u/30967867?s=20&v=4]JACKLiuDay left a comment (Joanna-HE/LIGO.#17)<#17 (comment)> I cannot understand “their timestamps don't seem to match". We definitely apply the time synchronization. The setting is for ligo system, the ligo doesn't use the online updated synchronized time for running, since 18s is accurate enough. This time difference, i.e., 18s, between gnss message and the LiDAR indicates that we synchronized those two sensors. 获取 Outlook for iOShttps://aka.ms/o0ukef … Sorry, Dr. He, I didn’t quite understand your point. If the GNSS data and LiDAR have already been hardware-synchronized, then why is there still an 18-second delay that needs to be set? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXUNVS3P7U4RD74PXVL22RQX7AVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJXGU2TIOJSGI>. You are receiving this because you commented.Message ID: ***@***.***>

[Joanna-HE]

I would like to suggest you to learn the information of gprmc, and the time baseline for the gps atomic time and the utc time. 获取 Outlook for iOS<https://aka.ms/o0ukef>

…

________________________________ 发件人: ***@***.*** ***@***.***> 发送时间: Monday, April 21, 2025 1:01:55 PM 收件人: Joanna-HE/LIGO. ***@***.***> 抄送: Comment ***@***.***> 主题: Re: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) Yes, this is the time difference between the gps time that included in the gnss message, and the utc time that is included in the gprmc message. The gprmc is sent to LiDAR for lidar time synchronization, giving lidar utc time, while gnss message uses gps time. 获取 Outlook for iOS<https://aka.ms/o0ukef>

________________________________ 发件人: Dayu ***@***.***> 发送时间: Monday, April 21, 2025 11:29:03 AM 收件人: Joanna-HE/LIGO. ***@***.***> 抄送: ***@***.*** ***@***.***>; Comment ***@***.***> 主题: Re: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) This is an external email. I cannot understand “their timestamps don't seem to match". We definitely apply the time synchronization. The setting is for ligo system, the ligo doesn't use the online updated synchronized time for running, since 18s is accurate enough. This time difference, i.e., 18s, between gnss message and the LiDAR indicates that we synchronized those two sensors. 获取 Outlook for iOShttps://aka.ms/o0ukef … Sorry, Dr. He, I didn’t quite understand your point. If the GNSS data and LiDAR have already been hardware-synchronized, then why is there still an 18-second delay that needs to be set? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXUNVS3P7U4RD74PXVL22RQX7AVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJXGU2TIOJSGI>. You are receiving this because you commented.Message ID: ***@***.***> [https://avatars.githubusercontent.com/u/30967867?s=20&v=4]JACKLiuDay left a comment (Joanna-HE/LIGO.#17)<#17 (comment)> I cannot understand “their timestamps don't seem to match". We definitely apply the time synchronization. The setting is for ligo system, the ligo doesn't use the online updated synchronized time for running, since 18s is accurate enough. This time difference, i.e., 18s, between gnss message and the LiDAR indicates that we synchronized those two sensors. 获取 Outlook for iOShttps://aka.ms/o0ukef … Sorry, Dr. He, I didn’t quite understand your point. If the GNSS data and LiDAR have already been hardware-synchronized, then why is there still an 18-second delay that needs to be set? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXUNVS3P7U4RD74PXVL22RQX7AVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJXGU2TIOJSGI>. You are receiving this because you commented.Message ID: ***@***.***>

[JACKLiuDay]

Yes, this is the time difference between the gps time that included in the gnss message, and the utc time that is included in the gprmc message. The gprmc is sent to LiDAR for lidar time synchronization, giving lidar utc time, while gnss message uses gps time.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Thanks a lot for your explanation, Dr. He. You must be a very patient and helpful person in real life—haha!

[JACKLiuDay]

I would like to suggest you to learn the information of gprmc, and the time baseline for the gps atomic time and the utc time.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Hello Dr., I purchased a GNSS module, the ZED-F9P, but I’ve been unable to successfully synchronize timing with the MID360. I’ve connected both the PPS line and the TXD line, but it still doesn’t show GPS synchronization. Have you encountered this issue before?

[Joanna-HE]

The frequency of the gprmc should be 1 Hz, not 10 Hz, this is a limit of mid360. 获取 Outlook for iOS<https://aka.ms/o0ukef>

…

________________________________ 发件人: Dayu ***@***.***> 发送时间: Sunday, April 27, 2025 11:08:36 AM 收件人: Joanna-HE/LIGO. ***@***.***> 抄送: ***@***.*** ***@***.***>; Comment ***@***.***> 主题: Re: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) This is an external email. [https://avatars.githubusercontent.com/u/30967867?s=20&v=4]JACKLiuDay left a comment (Joanna-HE/LIGO.#17)<#17 (comment)> I would like to suggest you to learn the information of gprmc, and the time baseline for the gps atomic time and the utc time. 获取 Outlook for iOShttps://aka.ms/o0ukef … Hello Dr., I purchased a GNSS module, the ZED-FP9, but I’ve been unable to successfully synchronize timing with the MID360. I’ve connected both the PPS line and the TXD line, but it still doesn’t show GPS synchronization. Have you encountered this issue before? ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXWAQDNJXVVHMTGEVLT23RC3JAVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMZSHEZDONBSGQ>. You are receiving this because you commented.Message ID: ***@***.***>

[JACKLiuDay]

The frequency of the gprmc should be 1 Hz, not 10 Hz, this is a limit of mid360.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Dr. He, I noticed that in your configuration file, the extrinsic parameters of the mid360 and GNSS don't seem to match the installation dimensions shown in the image. Are the extrinsic parameters of the mid360 and GNSS calculated based on the center of the top circle of the helical antenna?

[JACKLiuDay]

The frequency of the gprmc should be 1 Hz, not 10 Hz, this is a limit of mid360.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Dr. Hello, have you ever encountered a situation where the operating frequency of the Mid360 lidargradually decreases when synchronized with GNSS? It slowly drops from 10Hz to 1Hz over time. However, if I turn off the GNSS board, the frequency slowly rises back to 10Hz. It's a very strange phenomenon. The RMC message's synchronization monitoring shows a stable frequency of 1Hz

f5f9b4985b17eaf85ff37ab70b617272.mp4

[Joanna-HE]

I have not encountered this problem. Yes, it is weird, and I would suggest you to contact the official website of mid360. As another way, since the LIGO doesn't require strict time synchronization, you can calculate the time difference between lidar and gnss yourself, and give this value to LIGO. 获取 Outlook for iOS<https://aka.ms/o0ukef>

…

________________________________ 发件人: Dayu ***@***.***> 发送时间: Saturday, May 10, 2025 9:58:05 AM 收件人: Joanna-HE/LIGO. ***@***.***> 抄送: ***@***.*** ***@***.***>; Comment ***@***.***> 主题: Re: [Joanna-HE/LIGO.] How is data synchronization achieved between the lidar and the GNSS module? (Issue #17) This is an external email. [https://avatars.githubusercontent.com/u/30967867?s=20&v=4]JACKLiuDay left a comment (Joanna-HE/LIGO.#17)<#17 (comment)> The frequency of the gprmc should be 1 Hz, not 10 Hz, this is a limit of mid360. 获取 Outlook for iOShttps://aka.ms/o0ukef … Dr. Hello, have you ever encountered a situation where the operating frequency of the Mid360 lidargradually decreases when synchronized with GNSS? It slowly drops from 10Hz to 1Hz over time. However, if I turn off the GNSS board, the frequency slowly rises back to 10Hz. It's a very strange phenomenon. ― Reply to this email directly, view it on GitHub<#17 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANJUIXWE2Z3GNVYAIZFJMRL25VMK3AVCNFSM6AAAAAB3A2SFGWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQNRYGIYTONBTGY>. You are receiving this because you commented.Message ID: ***@***.***>

[JACKLiuDay]

I have not encountered this problem. Yes, it is weird, and I would suggest you to contact the official website of mid360. As another way, since the LIGO doesn't require strict time synchronization, you can calculate the time difference between lidar and gnss yourself, and give this value to LIGO.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Hello Dr.He, we have found the cause of the problem. The PPS terminal on the GNSS board has poor contact, which causes the PPS signal to become disordered when there is vibration. It works normally after we have made a jumper connection.

[JACKLiuDay]

I have not encountered this problem. Yes, it is weird, and I would suggest you to contact the official website of mid360. As another way, since the LIGO doesn't require strict time synchronization, you can calculate the time difference between lidar and gnss yourself, and give this value to LIGO.

获取 Outlook for iOShttps://aka.ms/o0ukef
…

Dr. He, I have successfully reproduced your code and recently read your paper and code. I have a question: GNSS can operate in RTK mode, so why didn’t you directly incorporate RTK as a factor constraint for fusion, but instead chose to process the raw GNSS observations to construct separate factors? Is it because RTK fixed solutions can sometimes be unreliable? Wouldn’t it be better to integrate differential RTK (e.g., Qianxun’s) into LIGO?