Precision navigation for lunar descent and landing

A detailed analysis of the algorithm for precision lunar descent and landing navigation is presented. A continuous-discrete dual-state extended Kalman filter is developed in which the position, velocity, and attitude of the vehicle constitute one state in the dual-state realization, and the selected landing site constitutes the second state in the dual-state realization. Concurrently,estimation of the errors associated with accelerometer and gyro based measurements, as well as the errors associated with external sensor measurements (such as altimetry and velocimetry), is performed. Additionally, deviations from the nominal position of the center of gravity of the vehicle with respect to the inertial measurement unit are estimated. It is shown via covariance analysis that the extended Kalman filter implementation is working properly, utilizing all measurements when available and extracting information to within the noise level of each sensor. Based on the results of the covariance analysis, it is shown that the position, velocity, and attitude of the spacecraft are observable. Additionally, it is shown that the biases of the accelerometer and the gyro as well as the map tie errors associated with the landing site are also observable. Contrarily, it is found that the sensor biases and the scale-factor and nonorthogonality errors associated with the accelerometer and the gyro are not observable.Aerospace Engineerin