Path Planning For Multiple Unmanned Aerial Vehicles Based On AOA Localization

Passive Location refers to the technology of not actively radiating electromagnetic waves to the reconnaissance target,but only relying on receiving the electromagnetic waves emitted by the target to achieve reconnaissance and localization of the target.Due to the absence of the need to emit electromagnetic signals,passive location has higher security and stronger concealment in the battlefield,and can complement active radar to form a more comprehensive reconnaissance system.Based on the principle of passive location,it can be seen that the relative layout of observation stations and targets in passive systems is an important factor affecting localization accuracy.When the measurement accuracy of the sensors is fixed,optimizing the layout of observation stations is the primary means to improve localization accuracy.In recent years,with the rapid development of drone technology,it has demonstrated flexible self-organization and rich scalability as an aerial platform,and the miniaturization of sensors and data link technology are becoming increasingly mature.Integrating the passive location system into drone clusters and utilizing the advantages of flexible deployment of drones have become a new approach to solve the deploy optimization problem of passive observation stations.Based on this idea,this article takes the Angel of Arrival(AOA)location system as an example to study the target position estimation,optimal station deploy,multi-UAVs path planning,and other issues involved in the process of achieving AOA localization by UAV clusters.A multi-UAVs path planning algorithm based on AOA localization is designed.The specific research content is as follows:1.Clarified the principles of two-dimensional and three-dimensional AOA localization,analyzed the factors affecting location errors,and explored the optimal station deploy within the determined area.Firstly,a detailed introduction was given to the principle of dual to multi stations AOA localization in 2D and 3D scenarios,and the corresponding Geometric Dilution of Precision(GDOP)was derived as an analysis indicator.The effects of variables such as angle measurement error,station location error,baseline length,and relative height on the positioning error within the region were simulated and analyzed.Secondly,considering the constantly changing distance between the UAV and the target,a signal measurement model with noise variance and distance correlation was constructed for a single radiation source.Taking into account the characteristics of localization solution in trajectory planning,and based on understanding the characteristics of various algorithms,a joint weighted maximum likelihood estimation and unscented Kalman filter position estimation algorithm was designed.By comparing with classical solution algorithms,the superiority of the joint algorithm in this paper was demonstrated.2.Firstly,considering the limiting factors in actual location scenarios,the stations deploy problem was transformed into an optimization model,and the PSO algorithm was used to simulate and solve the optimal station placement for 2,3,and 4 stations within the restricted area.The CRLB corresponding to distance related noise was derived,and based on this,the optimal station deploy principle for the target was analyzed.Secondly,the station layout evaluation indicators based on the Cramer Rao Lower Bound(CRLB)were elaborated,and the two-dimensional CRLB and Fisher information matrices were derived.Then,using minimizing the trace of CRLB and maximizing the determinant of Fisher information matrix(FIM)as indicators,the optimal station deploy in a twodimensional scenario was theoretically analyzed,and a pattern of unfixed optimal station layout and uneven separation angle was obtained.Furthermore,the optimal station deploy principle was interpreted from the perspective of the direction finding line,laying the foundation for the research in Chapter 5.3.A multi-UAVs path planning algorithm framework was constructed with the goal of improving AOA localization accuracy.Firstly,by integrating the distance related noise signal measurement model and position estimation algorithm proposed in Chapter 2,a multi-UAVs path planning model is established with the optimal heading angle combination of the UAV group as the decision variable,and the trajectory of minimizing CRLB derived in Chapter 3 as the objective function,taking into account practical constraints such as turning rate,maximum communication distance,and minimum safe distance between aircraft.Then,the penalty function is used to transform the constrained optimization problem into an unconstrained optimization problem.The PSO algorithm is used to solve the heading angle combination that guides the flight of the drone group,and the trajectory points are generated through cyclic iteration.Finally,simulation examples demonstrate that the trajectory generated by the proposed algorithm can effectively reduce location errors during flight,and can achieve optimal layout for the target while approaching it.In order to test the performance of the algorithm,gust noise jamming and no-fly zone jamming are set.Simulation results show that the proposed algorithm can still achieve effective route planning under gust jamming,and can automatically search the edge of the no-fly zone to form the optimal layout for the target.The research results of this article can be widely applied to passive location systems for motion platforms to improve localization accuracy and stations deployment efficiency,and adapt to the needs of information technology equipment in the era.