I have a projected data in northing, easting, elevation that came from a GPS receiver fitted on a Pipeline Inspection Gauge (PIG). (I do not have the raw data at all)
This PIG collected the GPS data by going through an old pipeline which is buried a few meters under the ground at most places. (There are no drawings showing the pipeline original alignment)
The northing, easting, elevation data looks fine, i.e. it plots as a nice line without any oscillations or zigzag pattern.
I have have been told to correct this data, i.e. to make sure that the alignment we see is the actual alignment of the pipeline underground.
I do not have any additional control point or Inertial Measurement Unit (IMU) data, i.e. the bends depicted by the GPS data may not be a true bend at all.
I have been told that you can use the Kalman filter to filter the northing, easting, elevation data.
So, my question WHY? Why do you want to filter the GPS data that has already been filtered inside the receiver (and probably with a kalman filter)?
Moreover, I have been arguing that we need additional information to check the accuracy of the data and the northing, easting, and elevation information ALONE cannot be used as a input for further filtering.