Research On Some Issues In Pulsar-based Navigation

With the rapid development in space exploration,much progress has been made in the navigation technology for spacecraft.The demand of navigation technology gets higher because of the spreading scope of exploration and the diversity of survey target.The currently common-used earth-based navigation systems,which suffer from the restrictions on accuracy,timeliness and cost,cannot fulfill the growing navigation demand.Pulsar-based navigation technique has attracted the attention of many researchers because of its characteristics of wide availability,independence high reliability and low cost.Given the background,this thesis studies the problems of pulse phase estimation and pulsar observation series designing in pulsar-based navigation system.The content of this thesis is summarized as the following:1.This thesis proposes a discrete Fourier transformation(DFT)-based method to estimate the pulse phase of pulsar's signal.The pulse phase of pulsar's signal is the key measurement in pulsar-based navigation.The accuracy in pulse phase estimation has large impact on the final accuracy of navigation.Further,the computational complexity of pulse phase estimation is also huge.The computational complexities of tradition methods of pulse phase estimation are analyzed in this thesis.Based on the time-shift property of DFT,this thesis suggests evaluating the delay in time domain by calculating the phase shift in frequency domain.In this method,only one certain item in DFT sequence needs to be calculated,thus the computational complexity is decreased.Gaussian function approximation and ?-function approximation are introduced to further reduce the computational complexity.Numerical simulations show that the proposed method achieves similar accuracy with that of traditional approaches and has less computational complexity.This method can improve the real-time performance of pulsar-based navigation system especially for on-broad implements.2.This thesis puts forward a method to improve the accuracy in pulse phase estimation based on sampling and weighted averaging.The traditional cross-correlation method is combined with the idea that "averaging multiple measurements reduces the random error in measurement".This method samples in the time tags of arriving photons and evaluates the pulse phase in each subset using cross-correlation method.Weighted averaging is performed among the results on the subsets to give a final estimation.Cross-correlation value is defined as the weight in weighted averaging.This thesis also analyzes the mechanism of this method and proposes two qualitative explanations:a signal noise ratio-based explanation and an error-difference trade-off explanation.Numerical simulations demonstrate that this method can improve the accuracy in pulse phase estimation.A discussion about the choice of sampling rate is also provided.3.This thesis proposes a method designing the observation strategy in pulsar-based navigation with probability ellipsoid.The pulsar-based navigation system usually operates as an orbit determination process.Dynamic model and measurement model are adopted in some optimal estimation method to estimate the position and velocity of the spacecraft.The covariance matrix is employed to measure the navigation accuracy,the concept of probability ellipsoid in statistical orbit determination theorem is utilized to describe geometric distribution of the error in orbit estimation,and a method choosing observation target is put forward by minimizing the size of posterior probability ellipsoid.According to three different ways of measuring the size of probability ellipsoid,three different strategies are designed:the first minimizes the length of posterior probability ellipsoid's largest principal axis,the second minimizes the sum of the squares of posterior probability ellipsoid's three principal axes,and the third minimizes the volume of posterior probability ellipsoid.The best strategy among them is chosen by a numerical simulation:minimizing the sum of the squares of posterior probability ellipsoid's three principal axes.Numerical simulations show that the accuracy in navigation results can be improved with a good strategy.In addition,as the size of probability ellipsoid has a property of "online computable" i.e.,it can be calculated before the real measurements are taken,the observation strategies can be designed during preliminary mission analysis.The new methods proposed in this thesis have the theoretical significance and the practical value of improving the accuracy and timeliness in pulsar-based navigation system.They also provide inspiration for future relevant studies.