水空两栖无人飞行器航姿测量技术研究

RESEARCH ON ATTITUDE AND HEADING MEASUREMENT TECHNOLOGY OF WATER-AIR AMPHIBIOUS UNMANNED AERIAL VEHICLE

冷沪思

11930431

硕士

材料工程

王凭慧

2021-05-19

2021-06-07

南方科技大学

深圳

近年来，空中飞行器技术日益成熟，为满足更多元化的应用场景，水空两栖飞行器逐渐成为人们研究的热点，它面临的环境具有一定复杂性，这对组合导航系统的可靠运行带来难度。在水下环境，水的电磁屏蔽作用导致卫星导航系统失效；空中飞行时，飞行器的高机动性以及环境干扰导致卫星信号发生跳变,本文针其水下工作的特殊性，进行了航姿模式的算法设计优化，为飞行器的水下控制提供依据；当其在空中飞行时，针对卫星信号易受干扰这一难题，进行组合导航模式的算法设计优化。 针对飞行器水下航姿工作模式，对航姿参考系统进行了误差四元数卡尔曼滤波器建模，以误差四元数，陀螺仪常值漂移误差，陀螺仪零偏估计误差作为9维状态量，以实际加计和磁力计输出与静态条件下它们的理论输出的差值，作为6维量测量。对误差四元数卡尔曼滤波器进行了优化，提出了基于新息的自适应卡尔曼滤波算法。对空中组合导航模式，依据捷联惯导误差方程，惯性传感器误差方程，设计了15维状态量的卡尔曼滤波状态方程，选择卫星导航与惯性导航的速度位置差值，以及惯性导航与磁力计解算航向角的差值作为量测量。提出了一种基于支持向量机的卡尔曼滤波算法，通过支持向量机和模糊逻辑器对算法进行改进。采集试验数据，验证了上述两种算法的可靠性。

In recent years, air vehicle technology has become more and more mature. water-air amphibious unmanned aerial vehicle has gradually become a research hotspot. The environment it faces has a certain complexity, which brings difficulties to the reliable operation of integrated navigation systems. In the underwater environment, the electromagnetic shielding effect of water causes the satellite navigation system to fail; when flying in the air, the high maneuverability of the aircraft and environmental interference cause the satellite signal to jump. The optimization of the algorithm design of the aircraft provides a basis for the underwater control of the aircraft; when it is flying in the air, the algorithm design optimization of the integrated navigation mode is carried out for the problem of the satellite signal being susceptible to interference. Aiming at the mode underwater, the error quaternion Kalman filter modeling of the attitude and heading reference system is carried out, and the error quaternion, the gyroscope constant drift error, and the gyroscope zero bias estimation error are regarded as the 9-dimensional state. The error quaternion Kalman filter is optimized, and an adaptive Kalman filter algorithm based on innovation is proposed. For the air integrated navigation mode, based on the strapdown inertial navigation error equation and the inertial sensor error equation, a 15-dimensional Kalman filter state equation is designed, and the speed position difference between satellite navigation and inertial navigation is selected, as well as inertial navigation and magnetometer Calculate the difference of the heading angle as a quantity measurement. A Kalman filter algorithm based on support vector machine is proposed, and the algorithm is improved through support vector machine and fuzzy logic. Collecting test data verifies the reliability of the above two algorithms.

水空两栖无人飞行器 卡尔曼滤波 新息自适应 支持向量机 模糊逻辑

water-air amphibious unmanned aerial vehicle kalman filter innovation adaptive support vector machines fuzzy logic

中文

独立培养

南方科技大学

冷沪思. 水空两栖无人飞行器航姿测量技术研究[D]. 深圳. 南方科技大学,2021.