The high-accuracy real-time navigation is crucial for pinpoint landing on the surface of the Mars. Most of the past and current navigation methods Most of the existing vision-based landing navigation algorithms assume that the maps of the Mars surface environments are known accurately in advance. However, the true situation is not the case. When a great change has taken place on the Mars surface reliefs during the photograph or the features prior known around the landing areas are sparse or non existed, for example, the mismatched features will increase. Moreover, the time-varying, strong nonlinear and uncertain characteristics of the Mars dynamics environments make the high-accuracy real-time navigation more difficulty. In this project, the landing navigation method in complicated and unknown Mars surface environment and limited observation condition is studied. Firstly the uncertainty propagation mechanism of the nonlinear landing trajectory with non-Gaussian uncertainties is studied, and the high-accuracy prediction method for the evolution of uncertainty based on the generalized polynomial chaos expansions is presented. Then the complementary navigation information provided by the vision and inertial sensors is fully investigated, the landing navigation observation scheme in the unmapped environment is established, and the landing navigation methods based on the inter-frame epipolar constraint and multi-frame geometry constraint is studied. In conclusion, this project aims to develop the new generation autonomous high-precision landing navigation method with strong adaptive capacity to the change of environment which is expected to provide a theoretical foundation and critical technical support for the future interplanetary landing missions.
高精度实时导航对于火星表面精确定点着陆至关重要,现有的光学着陆导航方法大多假定火星表面环境地图精确已知。实际情况并非如此,例如相机拍照时的火星表面地形地貌发生较大改变、着陆区域附近先验已知的特征点很稀疏或者不存在时,都会增加特征误匹配数目,再加上火星动力学环境的时变性、强非线性和不确定性,导致高精度实时导航极为困难。本课题针对复杂陌生的火星着陆环境和着陆器有限的观测条件,研究非线性非高斯着陆轨迹的不确定传播机理,提出基于广义多项式混沌扩展的着陆轨迹精确预报方法;充分挖掘光学-惯性敏感器提供的互补导航信息,建立无地图环境中的着陆导航观测方案,研究基于帧间极约束和多帧几何约束的着陆导航方法。本项目致力于发展新一代精度高、自主性强、环境适应能力强的着陆导航方法,对未来星际着陆任务的实施提供理论和关键技术支撑。
针对着陆器在陌生环境和有限观测条件下的导航问题,研究非线性光学测量模型选取策略,定量比较了焦平面测量模型和单位矢量测量模型的非线性强度,分析了测量模型非线性对导航性能的影响。充分挖掘光学-惯性敏感器提供的互补导航信息,建立无地图环境中的着陆导航观测方案,提出基于帧间极约束和多帧几何约束的着陆导航方法。应用数学仿真开展上述获得的模型和算法等的验证研究。该项目的研究成果为未来星际着陆任务的实施提供理论和关键技术支撑。该项目共发表论文7篇,其中EI 3篇;培养硕士研究生2名。
{{i.achievement_title}}
数据更新时间:2023-05-31
论大数据环境对情报学发展的影响
主控因素对异型头弹丸半侵彻金属靶深度的影响特性研究
中国参与全球价值链的环境效应分析
居住环境多维剥夺的地理识别及类型划分——以郑州主城区为例
基于协同表示的图嵌入鉴别分析在人脸识别中的应用
火星精确着陆自主导航策略与鲁棒估计方法研究
星际探测器精确着陆自主导航方法研究
大尺度未知环境下火星无人机自主导航方法研究
不精确先验地形条件下的AUV海底地形融合导航方法研究