基于联邦滤波的航迹推算方法研究

Research of Dead Reckoning Based on Federal Filter

摘要:

由于传统的基于IMU/里程仪组合的航迹推算系统受地形起伏及其打滑影响较大,影响了探测车导航系统的精度和稳健性,为了提高探测车导航系统的定位精度和稳健性,提出了一种基于联邦滤波的IMU/里程仪组合导航方法,该方法充分利用IMU提供的信息,在原航迹推算的方法上增加一个滤波器,将IMU的比力输出作为观测量进行滤波,并建立联邦组合导航系统的数学模型实现最优融合；从而提高了探测车导航系统的定位精度；最后地下矿道实验证明该方法能有效提高移动平台导航定位结果的精度和稳健性。

Abstract:

As the traditional dead reckoning system based on IMU and odometer is influenced by its undulating and slippery terrain, thus it affects the accuracy and robustness of the rover navigation system. In order to improve the positioning accuracy and robustness, the IMU / Odometer navigation system based on federal filter is proposed which takes full advantage of the information provided by IMU, adding a filter on the original method of dead reckoning which the output of ratio of power as the observations, then a mathematical model of the federal navigation system is set up to achieve the optimal fusion;Thus the high-precision positioning of the rover navigation system is achieved. Finally Underground Mine Road experiment proved the proposed method can effectively improve the accuracy and robustness of the positioning results of mobile platform navigation.