论文部分内容阅读
在光学辅助惯性导航系统中,观测信息的最优融合依赖于相机与惯性测量单元六自由度转换的精确校准。针对火星软着陆自主导航中的测量信息最优融合问题,提出了基于扩展卡尔曼滤波的导航敏感器相对位姿校准算法。该算法仅利用火星表面可获取的路标特征点信息,不借助额外的测量设备,对相机与惯性测量单元相对位姿进行精确的校准,同时,能够估计着陆器的位置、速度和姿态。考虑到着陆器机动和火星自旋的影响,建立了宽视场相机及惯性测量单元的高精度测量模型。最后通过数学仿真对所提出的校准算法的可行性和有效性进行了验证。
In the optical assisted inertial navigation system, the optimal fusion of the observed information relies on the precise calibration of the six degrees of freedom conversion of the camera and the inertial measurement unit. Aiming at the optimal fusion of the measurement information in Mars soft landing autonomous navigation, a relative position and attitude calibration algorithm of navigation sensor based on extended Kalman filter is proposed. The algorithm uses only the information of the roadmap feature points available on the surface of Mars, and precisely calibrates the relative pose of the camera and the inertial measurement unit without the aid of additional measurement equipment. At the same time, it can estimate the position, velocity and attitude of the lander. Considering the influence of lander maneuver and Mars spin, a high precision measurement model of wide field of view camera and inertial measurement unit was established. Finally, the feasibility and validity of the proposed calibration algorithm are verified through mathematical simulation.