Innovative Navigation Guidance And Control Technology For Mars Entry Descent And Landing

The current Mars missions still adopt the first generation of Mars EDL navigation, guidance and control technology which were developed during the Viking mission. Viking-like landers adopted the dead reckoning (DR) based navigation mode and unguided ballistic trajectory entry, which lead to larger landing error ellipses in the order of several hundred kilometers and cannot meet the requirements of future Mars landing and sample return missions. Therefore, it is necessary to develop a new generation of Mars EDL navigation, guidance and control technology.In this dissertation, with precision landing on Mars exploration mission as the engineering background, the next generation of Mars EDL navigation, guidance and control technologies are studyed from the following three aspects: the high-precision EDL navigation, trajectory optimization and guidance control. The main contents of this dissertation include:First, the dynamic model of entry, descent and landing phase are built up by reasonable assumptions and approximations in order to obtain the the dynamic characteristics of the probe, which provides the necessary mathematical model for the following guidance and control system design.Second, the external measuring sensors/IMU integrated navigation algorithms for Mars EDL are developed to overcome the issue of navigation accuracy low of traditional IMU based dead reckoning due to the error accumulated over time. Radio measurement and communication between probe and orbitor/beacons are used to correct the inertial constant bias during entry, simulation results show the navigation accuracy can be significantly improved and is not affected by radio measurement interrupt. In descent phase, IMU and miniature coherent altimeter and velocimeter are combined for integrated navigation. Although the horizontal position is not observable, the integrated navigation scheme overcomes the shortcomings of inertial navigation error divergence. The navigation accuracy requirement of landing phase is higher, therefore the navigation camera is included to better estimate the horizontal movement.Then, the nominal trajectory of entry and landing phases are pre-planned. Here, the limited control ability of the vehicle are taken into account in the trajectory optimization process, which effectively avoid the undesired situation that the vehicle can not track the nominal trajectory due to control saturation when the true trajectory is much different from the nominal trajectory. At the same time, the sensitivity index of the state variables with respect to uncertainty is included to optimization performance index of the nominal trajectory with almost not changing the original performance index, which can effectively reduce uncertainty effects and then improve the control precision.Finally, an adaptive guidance law is proposed according to CGT-based direct adaptive control theory in order to effectively eliminate the adverse effects of uncertainty parameters on the guidance and control system. A sliding-mode variable structure controller is also designed to overcome the problem that the entry dynamic model includes the unmodeled dynamics, and then the uncertainty terms in variable structure controller are online estimated using neural networks, so the guidance control system has strong robustness.