自适应联邦卡尔曼滤波在机器人组合导航系统中的应用研究

利用里程计(OD)与全球定位系统(GPS)辅助捷联惯性导航系统(SINS)构成一种高可靠性的组合导航系统.推导并建立了局部滤波器的数学模型,并针对联邦滤波器在载体发生异常扰动时滤波精度较低的问题,设计了基于SINS/GPS/OD组合导航系统的自适应联邦滤波器,有效补偿了系统异常扰动或动力学模型误差.仿真模拟了机器人的全航线运行轨迹进行验证,仿真结果表明,SINS/GPS/OD组合导航系统的自适应联邦卡尔曼滤波算法与相同组合导航系统的非自适应联邦卡尔曼滤波算法相比,在保障机器人导航定位可靠性及容错能力的前提下,能有效抑制异常扰动的影响,导航精度得到进一步改善.

Application of Adaptive Federated Kalman Filter in Robot Integrated Navigation System

The odometer (OD) and global positioning system (GPS) are used to aid strapdown inertial navigation system (SINS) to constitute an integrated navigation system with high reliablity.A mathematical model of local filter is established.In order to solve the problem that the federated filter''s precision can not meet demand when the carrier has an abnormal disturbance,the adaptive filter based on the SINS/GPS/OD is designed which can effectively compensate abnormal disturbance or dynamic model error.The whole line trajectory of the robot is simulated,the results show that compared with the non-adaptive,the adaptive federated Kalman filter algorithm based on SINS/GPS/OD integrated navigation system can effectively suppress the effect of abnormal disturbance and further improve the navigation accuracy,under the premise of ensuring the reliability and fault tolerance of robot navigation.