| In recent years,the integrated manufacturing technology of UAVs has been popularized rapidly,and the components are becoming miniaturized,low-cost and low-energy consumption.Restricted by price factors,low-cost sensor navigation has become the main application requirements in the UAV market.UAVs often operate in mountainous forests,where the environment is complex and changeable,and the navigation process is highly maneuverable.With the continuous improvement of remote navigation performance requirements,a single navigation mode can not meet the needs of actual tasks.Therefore,modern UAVs mostly adopt the form of multi-sensor integrated navigation,and each system has complementary advantages to ensure the robustness of the system.Firstly,the error models and basic working principle of various low-cost sensors are studied;Then,based on the analysis of the error models of Inertial Navigation System(INS),Global Navigation Satellite System(GNSS),barometer and magnetometer,the state equation and measurement equation are established,and the Kalman filter is used to realize the calculation of multi-sensor integrated navigation.Due to the instability of traditional Kalman filter algorithm under high dynamic conditions,an adaptive filter algorithm with improved adjustment factor is proposed.The algorithm uses the three-dimensional velocity information of Global Positioning System(GPS)to optimize the adjustment factors in the filter anomaly determination conditions in real time,dynamically adjust the filter anomaly judgment nodes,estimate the measurement noise covariance matrix in real time,and improve the self adaptability of the system;Finally,in order to prove the feasibility of the improved algorithm in practical application,the navigation sensor is used for field data acquisition and off-line simulation.The simulation results prove the feasibility of the improved algorithm in the application of measured data. |
