In this paper, the UAV formation position adjustment is divided into two steps: transmitting signal UAV position adjustment and receiving signal UAV position adjustment. Using geometric linear motion, the position adjustment of the transmitting signal drone is achieved without knowing the distance in advance. Subsequently, in order to exclude the influence of environmental factors (wind speed, etc.) on the flight direction of the UAV, a pure orientation passive positioning and navigation model based on Kalman filtering was established, and the UAV flight command vector was corrected using filtering techniques, which enabled the model to adapt itself to the environment for UAV navigation. In order to adapt to the topic specificity, the ReLU function is introduced in the state estimation function to further ensure the stability of the correction vector. After the model's navigation, the UAV is finally able to stabilize near the target location with centimeter-level accuracy. The differences between conical and circular formations are discussed, and the geometric positioning model is modified to avoid the special problems caused by the specificity of conical formations as much as possible. The initial state of the formation is randomly generated, and the UAV position is adjusted using a pure azimuthal passive positioning navigation model. The results show that the UAV positioning is still able to maintain centimeter-level accuracy.
Research on pure azimuth passive positioning and navigation problem of UAV based on Kalman filter
26.05.2023
1834986 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch