Is it possible to use multiple base station in ardupilot? if yes how? and also how far can rtk correction data will work perfectly for rover?
Do you know the meaning of Ardupilot Swag?
No, I don’t have any idea about that. what is it? and how it works?
This thread is for Ardupilot T-shirt’s, Coffee Mugs, things like that. Nothing at all related to what you posted. Choose the Appropriate Rover thread.
1 Like
Yes we developed something similar. Because a drone accuracy for rtk will get further ever 10km. Although not much feeding from multiple location is better. But we used python and wrote our own injection. So… Not sure how to do it in gcs software
The answer is:
It is possible if you write your own RTCM multiplexer software.
I have.
I can’t say it was any better or worse.
I have supplied RTCM to the GPS directly from one source, and via MavLink to the GPS via another source.
I can’t say which was used for what, or what impact it had since I never looked “inside” the F9P to see how it felt about two sets of corrections coming in.
But it worked and I got a FIX.
But using just one correction does the same.
That’s not a good plan. Your GPS module can only accept corrections from one source at a time, so sending multiple simultaneous sources only has the potential to hurt, not help.
I think the intent of the OP was to be able to switch sources based on proximity to a base station, which is not directly supported but could be managed manually.
It may also be possible to collect data from multiple nearby stations, process it, and obtain a higher degree of precision than any one station would provide, but you’d need a custom software solution to do that and output a single correction stream for use with the vehicle.
1 Like
This is exactly what RTN services do, they have a fancy software that calculates a virtual base station near a position you asked for, Mission Planner supports such stations.
Absolutely. But you can’t just go connecting various sources all nimbly bimbly hopping from tree to tree and expect good results like the previous user suggested.
2 Likes
To be clear, I didn’t suggest good results.
I specifically said I didn’t think it was any better than a single source, and it was impossible to know which base was used - or what impact two bases has - without knowing what was going on inside the GPS.
I.e. it works, but I wouldn’t recommend it.
I didn’t even do it intentionally - I was trying to sort out some communications problems and differences between using NTRIP vs a direct TCP and what filters got applied where, and even using UBX protocol and wound up feeding corrections from two disparate sources (not just two forms of the same base) on accident.
It worked.
I don’t think it worked better than one base.
I didn’t evaluate if it was more or less accurate.
I only know, and stated, it came back FIXed.
Meanwhile, I resolved my tests to make sure NTRIP vs TCP and RTCM vs UBX and went back to one NTRIP sourced RTCM feed from one base.
I don’t see the value in supplying corrections from two sources at once, particularly without knowing how the GPS processes that, but it can be done.
I assumed the OP had a specific use case in mind.
There is no value, and it should not be done.