Development of Optimal Guidance and Adaptive Algorithm for Vertical Landing Vehicle

1. a）从2002年到2003年，研究了月球着陆器垂直和软着陆的最佳制导律，并在2002年和2003年在韩国举行的ICCAS会议上介绍了所获得的结果。B）假设月球着陆器的运动方程是线性的，从理论上推导出了一种新的最小加加速度最优制导律，并于2004年将其仿真结果提交给了日本航空航天学会学报。本刊2005年第48卷第159期将对此进行专门论述. c）2004年，假设再入飞行器的运动是非线性的，利用先前获得的最小加加速度最优制导律，使用精确线性化方法，开发了一种新的制导律。研究结果已提交给《日本航空宇宙学会杂志》，并将于2004年夏季发表。2.多DSP实时仿真 关于我们 boardsa）为了验证所提出的最优制导律的有效性，在多DSP板上进行了真实的实时仿真。在仿真中，考察了有限字长、离散化引起的限零问题、计算时延和测量噪声对控制系统的影响。B）特别是对限零问题进行了理论分析，阐明了采样时间与限零问题之间的关系。通过回顾提取过程来进行更精确的模拟，在提取过程中，从关于剩余时间的六次代数方程获得的真实的数值解中选择最佳剩余时间解。d）最后，在考虑实际传感器规格的模拟中，发现测量噪声导致终点（目标点）的下靶场误差最多偏离12 m;但其对油耗的影响可以忽略不计。这一结果于2004年3月在日本航空航天科学学会的一次国内会议上提出。少

1.Development of optimal guidance lawa)From 2002 to 2003, an optimal guidance law was investigated for the vertical and soft landing of a lunar lander, and obtained results were presented at the ICCAS conferences held in Korea in 2002 and 2003. b)A novel minimum jerk optimal guidance law was theoretically developed by assuming that the equation of motion of the Lunar lander was linear, and was submitted to the Transactions of the Japan Society for Aeronautical and Space Sciences with its simulation results in 2004. An article on this will appear in Vol.48,No.159 of this transaction in 2005. c)In 2004, by assuming that the motion of a reentry vehicle was nonlinear, a new guidance law was developed using an exact linearization method with the previously obtained minimum jerk optimal guidance law. The result was also submitted to the Journal of the Japan Society for Aeronautical and Space Sciences and will appear in this journal in the summer of 2004.2.Real time simulation using muti-DSP … More boardsa)A real time simulation was carried out using multi-DSP boards in order to confirm the validity of the above- mentioned optimal guidance law. In this simulation, the effects of finite-word length, a limiting-zero problem due to discretization, computational time delay, and measurement noise on the control system were examined, b)In particular, a theoretical analysis of the limiting-zero problem was carried and the relationship between the sampling time and the limiting-zero problem was clarified, c)Furthermore, a more precise simulation was carried out by reviewing the extraction process in which an optimal time-to-go solution was chosen from the real number solutions obtained from a six-degree algebraic equation with respect to time-to-go. d)Lastly, in the simulation in which actual sensor specifications were considered, the measurement noise was revealed to cause the downrange error to deviate at most 12 m at the terminal point (target point) ; however, its effect on fuel consumption was negligible. This result was presented at a domestic conference of the Japan Society for Aeronautical and Space Sciences in March, 2004. Less