2 mobile beacon timestamp synchronization

Frequent questions asked online, offline, in forums are answered here
Post Reply
ericchio
Posts: 2
Joined: Fri Jan 27, 2023 10:38 am

2 mobile beacon timestamp synchronization

Post by ericchio »

Hello, I want to get the localization/orientation on the DJI M3 Drone, so I put:
1. four stationary beacons to build a simple submap
2. two mobile beacons on the top of the Drone to get the center position and the orientation
3. Beacon version: HW v4.9
4. NIA mode
5. use ROS to get the data

And I found that the data is output one by one, resulting in inaccurate (even trembling) estimated orientation. I would like to know if there is any way to synchronize the timestamps of the two beacons, I think this should greatly solve the problem of inaccurate attitude.

Thanks

User avatar
admin
Site Admin
Posts: 393
Joined: Tue Jan 26, 2016 6:06 pm
Contact:

Re: 2 mobile beacon timestamp synchronization

Post by admin »

Hello,

Right. In NIA, only one mobile beacon can be served at a time. Since you have Beacons HW v4.9, I assume you use them as stationary beacons as well. In this case, you can use only NIA.

With Super-Beacons, it is possible to use MF NIA and emit from both Super-Beacons at the same time because they will emit on different ultrasound frequencies and stationary Super-Beacons will be able to receive both frequencies at the same time. Thus, there will be no "left-right, left-right effect".

Besides, Super-Beacons support Ultrasound+IMU sensor fusion for direction. You get a very accurate direction with no drift.

Also, please, check and follow the logic about pairing the beacons. Maybe, the problem is simpler:
- Achieve perfect tracking of mobile beacon 1 - independently of Beacon 2. Just switch the Beacon 2 off completely
- Achieve perfect tracking of mobile Beacon 2 - the same way as above - without the Beacon 1 active
- Then, achieve perfect tracking of them simultaneously, but without pairing
- And only then pair them together and achieve perfect tracking with Location + Direction

Make sure you are not confused with the distance between beacons settings in the Dashboard - it is between the centers of the beacons.

BR,
Maxim

ericchio
Posts: 2
Joined: Fri Jan 27, 2023 10:38 am

Re: 2 mobile beacon timestamp synchronization

Post by ericchio »

Thanks for your reply.
Here are three more questions I would want to know.
1. In MF NIA, will the system automatically schedule the two hedges to emit sound waves at the same time? Any additional settings required compared to pure NIA?
2. Since I put the two hedges on the DJI Drone, so their heights will keep changing and not necessarily parallel to the XY plane during flying. Can I still use your pair function on Dashboard?
3. I put the two beacons on the top of the DJI Drone, the max baseline of 2 hedges I can set is about 12cm. I think it is a big problem to get the accurate yaw angle because the error of the beacons is +-2cm, theoretically the max angle error is atan(4/12) which is about 20 degrees. Can the IMU fusion help in this situation?

Thank you and hope for your reply again.

User avatar
admin
Site Admin
Posts: 393
Joined: Tue Jan 26, 2016 6:06 pm
Contact:

Re: 2 mobile beacon timestamp synchronization

Post by admin »

ericchio wrote:
Mon Feb 13, 2023 7:02 am
Thanks for your reply.
Here are three more questions I would want to know.
1. In MF NIA, will the system automatically schedule the two hedges to emit sound waves at the same time? Any additional settings required compared to pure NIA?
2. Since I put the two hedges on the DJI Drone, so their heights will keep changing and not necessarily parallel to the XY plane during flying. Can I still use your pair function on Dashboard?
3. I put the two beacons on the top of the DJI Drone, the max baseline of 2 hedges I can set is about 12cm. I think it is a big problem to get the accurate yaw angle because the error of the beacons is +-2cm, theoretically the max angle error is atan(4/12) which is about 20 degrees. Can the IMU fusion help in this situation?

Thank you and hope for your reply again.


1) Everything is done fully automatically. You don't have to do anything. Effectively, you just enjoy up to 8 times higher update rate per mobile beacon as compared with NIA, because there will be up to 8 simultaneously emitting mobile beacons on 8 different ultrasound frequencies

2) Yes, because for the drones you will be using 3D tracking most likely. Thus, you don't enter the heights for the mobile beacons at all. They will be measured by the system. The heights for the stationary beacons must still be provided

3) Yes, 12 cm is a small base. People did with such a base, it worked somehow, but it was not perfect. Still, since there is an IMU+Ultrasound sensor fusion, which help a lot, typical performance is satisfactory in the majority of cases

Please, don't hesitate to ask questions. We are always happy to explain and clarify.

Post Reply