摘要
基于自适应神经模糊逻辑推理系统(ANFIS),在全球定位系统(GPS)信号阻塞时,为惯性导航系统(INS)提供位置和速度修正量以提高系统的精度和鲁棒性.首先用小波对数据信号进行降噪处理;然后设定INS的位置或速度作为ANFIS的输入参数,经训练后输出相应修正量,训练期望值为经小波多分辨率分析得到的位置误差和速度误差.实验表明,无GPS信号时定位精度比同条件下卡尔曼滤波精度提高约40%,因此该方法可为车辆提供可靠有效的导航定位服务.
Based on adaptive neuro-fuzzy inference system(ANFIS),corrections of position and velocity are provided for inertial navigation system(INS) when global position system(GPS) signal is blocked to improve the accuracy and robustness of the whole system. To achieve the object,wavelet is used to de-noise the output of the sensors,then positions and velocities are set to be the input of ANFIS,while position and velocity corrections are output after trained. The desired outputs of trained samples are the position and velocity error after wave multi-resolution analysis. The result shows that,when GPS is blocked,position accuracy is increased forty percent than Kalman filtering,and this method can provide reliable and effective positioning services for vehicles.
出处
《控制与决策》
EI
CSCD
北大核心
2010年第7期1109-1112,共4页
Control and Decision
基金
中国博士后基金项目(20080441012)
国家自然科学基金项目(60904088)
总装备部预研项目(51309060402
51309020503)
陕西省电子信息系统综合集成重点实验室基金项目(200909A)
东南大学科技基金项目(KJ2009382)