Cooperative Navigation Algorithm Of UAV Formation Based On Multi-sensor Fusion

In recent years,with the continuous development and maturity of unmanned aerial vehicle(UAV)technology,the application of UAV has gradually developed in the direction of formation and clustering.Various countries have gradually carried out research on UAV formation.High-precision navigation information is the prerequisite for realizing the functions of the UAV formation,and the single navigation sensor in the UAV formation navigation system has certain defects.It cannot meet the needs of UAV formation navigation.In order to improve the overall navigation accuracy of the UAV group and meet the needs of the UAV group combat function,the UAV formation cooperative navigation algorithm and the tracking of enemy UAV targets algorithm are studied in this article.The main work of this article is as follows:Firstly,the algorithms and mathematical models of related navigation methods are studied.Strapdown Inertial Navigation System(SINS),Beidou Satellite Navigation System(BDS)solution algorithm,air target visual positioning algorithm and radar positioning algorithm are studied.The error equation is derived,the corresponding mathematical model is established and simulated.The basis is provided for the collaborative navigation of the master-slave UAV formation and the target tracking during UAV group operations.Secondly,collaborative navigation algorithm for UAV formation is studied.The BDS/SINS integrated navigation system is used to realize the correction of the lead UAV inertial navigation,and the integrated navigation filter is designed according to the error model.Then,aiming at the problem of low navigation accuracy of the wingman in the UAV formation,a collaborative navigation method using the relative navigation information between the leader and the wingman is proposed.The relative navigation information with the lead UAV is obtained through the laser ranging,angle measurement sensor and Doppler speed sensor carried by the wingman,and it is introduced into the observation equation of the filter to correct the navigation error of the wingman’s inertial navigation.The simulation results show that this method can effectively realize the collaborative navigation of the UAV formation.Finally,aiming at the problem of low tracking accuracy due to different target motion states when the UAV formation cooperates to track the enemy UAV formation,a vision/radar-based UAV formation interactive multi-model multi-target tracking algorithm is proposed.According to the common target motion model,a vision/radar information fusion filter is designed that introduces the pixel coordinate information of the moving target into the observation equation to improve the accuracy of target tracking.A vision/radar-based interactive multi-model multi-target tracking algorithm is proposed based on this filter,and the interactive multi-model algorithm(IMM)and track correlation algorithm are applied to the tracking of enemy UAV formation.The simulation results show that when different targets in the enemy UAV formation are maneuvering,the algorithm can also track the targets.