To What Extent Can Standard GNSS Ambiguity Resolution Methods Be Used For Single-Frequency Epoch-by-Epoch Attitude Determination?

摘要

In this contribution we study and compare the strength of three different GNSS models for carrier-phase attitude determination.These models are(1)the unconstrained GNSS model,(2)a linear(ized)constrained GNSS model, and(3)a nonlinear constrained GNSS model.The latter model is the baseline-length constrained GNSS model, while the second model is its linearized version.The strongest of the three models is the third.The advan- tage of the first model is,however,that one can apply any standard ambiguity resolution(AR)method to it.This is not true for the third model,as special precautions need to be taken in order to rigorously incorporate the nonlin- ear constraints.The strength of the second model ap- proaches that of the third if the nonlinearity is negligi- ble.The constrained LAMBDA method has been devel- oped for the third model,while the standard LAMBDA method can be directly applied to the first and the second model.We show the preferred method for different mea- surement scenarios and different baseline lengths.With our results,a user will be able to determine for which baseline length he/she needs to switch from the standard LAMBDA method to the more advanced constrained ver- sion.In our results we emphasize the most challenging case,which is the single-frequency,epoch-by-epoch,un- aided GNSS attitude determination.