MC-ROV导航系统研究  

Research of MC-ROV Navigation System

作　　者：吴长进 刘慧婷[2] 

机构地区：[1]扬子江船业集团公司,江苏靖江214500 [2]江苏科技大学电子信息学院,江苏镇江212003

出　　处：《计算机测量与控制》2017年第4期159-161,共3页Computer Measurement &Control

摘　　要：针对研制的新型模态切换水下机器人(Mode—Converted ROV,MC—ROV),设计了一套以MEMS器件为主的微惯性导航系统,包括三轴陀螺仪、三轴加速度计、三轴磁力计及高精度深度传感器;基于加速度计和陀螺仪的频域互补原理,系统首先运用互补滤波算法融合传感器数据,补偿陀螺漂移,提高系统动态性能;然后分析了四元数微分方程,以四元数估算水下机器人的角度大小和方向,可显著减小计算量;最后采用了改进的渐消记忆指数加权自适用卡尔曼滤波器,增大新近数据的作用,相应减小陈旧数据的作用,有效避免滤波发散,提高导航精度;水池实验表明,在持续震动和电磁干扰等恶劣环境下,利用互补滤波和改进的自适应卡尔曼滤波,水下机器人依然能够获得准确的、高可靠性的水下姿态信息。For the novelty Mode-Converted ROV （MC-ROV）, a set of MEMS-based integrated inertial navigation system has been developed, including 3-axis gyroscope, 3-axis accelerometer, 3-axis magnetic compass and high-precision depth sensor. In terms of the compensation infrequency domain between accelerometer and gyroscope, the system firstly utilizes complementary filter to merge the sensors in order to restrain the drifts of gyroscope and improve the dynamic performence. Then quaternions ＇ differential equation is analised and further to be the estimated values of MC-ROV＇ s position and angles, which apparentely reduce the computing. This paper applies an i proved fading memory index--weighted adaptive Kalman filter that stresses the effects of new data while weakening that of old data to avoid divergence and further enhance the navigation precision. The pool experimental results present that using complementary filter and the im proved Kaiman filter can aquire underwater angles with high precision and reliability, though in harsh ambience such as continuous vibration and electromagnet interference.

关 键 词：水下机器人 微惯性导航 四元数 自适应卡尔曼 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]