RTK Verification

[Nistorescu Sorin]

Hi,

Although J-Field become so advanced regarding the statistical methods that provide confidence and consistency, in some canopy isolate points, we need more than 2 steps + validate rover algorithm.

The base can also play an important role by taking the final decision in this process.

So, in practice, only for some difficult points, we usually invert the base and rover. We keep the rover in a stable position with a tripod clamp and apply a base profile. Then apply a rover profile to the base. Finally we can compare the vector and coordinate to the main base like in a classical two way traverse.

For us this is the final step before validating a RTK position using one of the advantages of Javad receivers to be rover or base.

We know that the base should not be use in bad places, but for some isolated locations this procedure give more confidence. Note that in dense canopy, the inclusion of BEIDOU observations into RTK can improve the position availability by up to 30%. After inversion, the original base-rover profile can be restored.

Very useful could be an additional display system at the rover using RAMS, to remote control the base (usually android phone/tablet). Otherwise we need a second person to change the base profile. This way of working is more suitable in areas with good coverage of GSM carriers and with TCP IP corrections.

Regards.

 

[Shawn Billings]

Interesting theory. So you perform base to rover vector A-B. Then set rover to be a base and base to be a rover and determine the vector from B-A.

Is it significantly better than repeating vector A-B?

 

[Nistorescu Sorin]

Shawn,

Sometime it is like a safety belt. We look to return for a satisfactory value close enough to the main base coordinate. If our statistic strategy regarding vector A-B is good then the two-way coordinate finally obtained in A will be close from the original.

Repeating vector A-B give us confidence and consistency but inverting the system will ensure that our RTK measurement for B coordinates do not exceed usual errors. Consider it only like one last check to avoid a bad result. In the verification process all the methods will count to a final good solution. Nothing is significantly better or worse.

Without lose our heads in statistics, I would like to say that we will be closer to the initial base coordinate as it will develop all segments of current GNSS systems - ground, space, civil.

As an small example look at TRE-3 OEM Board specifications: highly stable digital filters (band characteristics do not change with age, input voltages, or temperature) improved GLONASS interchannel bias performance, excellent new multipath rejection technique, low signal strength reception under-canopy, etc.

So, do a usually both base and rover system match the exigence? Even if they match there is an ongoing update process for the GNSS ground control stations, satellites and techniques.

Some aditional tests must be done but we will always try not to force the system in very bad canopy because it was not designed for this purpose. And always keep in mind the quality of the equipment.

Regards.

 

[Nistorescu Sorin]

The results of such a method can be both simple or spectacular. A quality evaluation system was implemented in our area.
http://monitoringeupos.gku.sk/

Personally, I see DPOS not only a very good instrument for coordinates determination, but also an important tool for critical points verification, using two-way vector processing and main base coordinate comparison.