High precision positioning and navigation technology for unmanned aerial vehiclesin high-risk environments

Abstract:In high-risk and complex mountainous environmental conditions, geological hazard bodies are often located in high mountain and canyon areas, with problems such as poor observation environment, large altitude differences, frequent extreme weather, and poor data communication. When deploying GNSS monitoring equipment using drones, they face difficulties in safe flight and precise navigation and positioning.This article studies the performance enhancement technology of multi-sensor integrated navigation. Based on the low-cost MEMS and GNSS tight combination algorithm, a comprehensive filtering technology integrating unscented Kalman filtering(UKF) and particle filtering(PF) is proposed. Two parallel running UKF and PF predictors are used to alternately use the information received by the fusion filter through input interaction, model filtering, and adaptive filtering optimization, in order to improve the reliability and accuracy of filtering, Simultaneously combining the advantages of Fuzzy Inference System (FIS) and Sparse Random Gaussian Model (SRG), the SRG model is used to estimate the initial state vector, and then FIS is used to update all current states to obtain the optimal predicted state vector. During the interruption of GNSS signal, the measurement data trained by FIS is jointly provided by INS and GNSS. In order to improve the required prediction accuracy, time sliding is used to control the data flow generated by INS and GNSS, Fully utilize the linear velocity and angular velocity increments output by INS to update the attitude, velocity, and position of unmanned aerial vehicles, improve the filtering convergence speed, navigation positioning accuracy, and reliability in the event of GNSS signal interruption. This article designs a multi-source fusion positioning system framework and integration scheme for unmanned aerial vehicle applications, and develops a high-precision fusion positioning software system, mainly including GNSS real-time data reception, GNSS data preprocessing, GNSS/INS integrated navigation data processing, real-time positioning result forwarding and other modules. The test results show that the positioning accuracy of complex environment integrated navigation is better than 0.5 meters. Key words:Geological disasters;Unmanned Aerial Vehicle;Integrated navigation;High precision