Research On Intelligent Control Technology Of Multi-UAV Cooperative Radar Jamming

With the development of radar science and technology,traditional jamming technology faces enormous challenges.And to achieve a better jamming effect,multi-jammer cooperative jamming technology is gradually developed.Unmanned Aerial Vehicles(UAVs)can perform missions in hazardous environments instead of human-crewed aircrafts,making them an ideal carrier for jamming.In addition,multiple UAVs working in tandem can perform more complex tasks than a single UAV.Therefore,the use of multiple UAVs for cooperative jamming is an effective way.This thesis focuses on the military requirements of multi-UAV cooperative radar jamming,starting from the intelligent control of the whole process from flight control,resource management to jamming decision-making,and proposes corresponding solutions to the problems of existing technologies to improve the cooperative jamming effect.The main work and innovation points of this thesis are as follows:1.The multi-UAV cooperative radar jamming research is relatively new,and there is a lack of dedicated methods for flight path planning in this scenario.The existing flight path planning methods do not consider the characteristics of the formation and its environment,so the effect of their migration to this scenario is not good.Therefore,in this thesis,from the five aspects of flight path length,flight path feasibility,flight path safety,cooperation constraints among aircrafts,and mission accomplishment effect,a flight path planning model for multi-UAV cooperative radar jamming is constructed by combining the characteristics of the formation and its environment.In response to the problems in the existing flight path planning methods,to effectively describe the maneuver properties of each aircraft and the connection between the accompanying jammer and the target aircraft,a Multi-Spherical vector-based(MS)method is proposed.At the same time,to avoid mutual interference of flight path point information,a Hybrid Particle Swarm Optimization with Fight path point by Flight path point Learning capability(FLHPSO)optimization algorithm is proposed,and the two algorithms are combined to form the MS-FLHPSO model solving method.The simulation results show that,in comparing the mean values of the optimal solutions for different scenarios,and the average improvement of the MS method is more than 30%,the FLHPSO improves more than 10% on average compared with the HIPSO-MSOS.Moreover,the algorithm in this thesis is superior to other algorithms in terms of the remaining indexes of the optimal solutions.It fully demonstrates that the algorithm in this thesis can plan flight paths with higher reliability under ensuring stability.2.Based on the completed flight path planning,a Combination Search Strategy(CSS)-Improved Particle Swarm Optimization(IPSO)based multi-jammer cooperative resource allocation method is proposed in this thesis to improve the utilization rate of jamming resource further.The existing multi-jammer resource joint optimization method adopts the two-step solution,which usually only optimizes the sub-problems,leading to low utilization of jamming resource.This thesis proposes the CSS method to address this problem.The method can directly find the global solution to the original problem by constructing the corresponding decision variables to describe the combined relationship between beam pointing and transmit power.In addition,since the CSS method leads to an increase in the complexity of the decision variables,to further improve the optimization performance of the PSO algorithm,we introduce a self-correction strategy while proposing a variable-by-variable learning strategy to obtain the IPSO algorithm.The simulation results show that,compared with the two-step solution method,the single experiment time of CSS is reduced by about 1.5s on average,and about 8 percentage points reduce the detection probability on average.It fully demonstrates that the CSS method can obtain better resource allocation results in a shorter time.Meanwhile,IPSO can further ensure the stability of the resource allocation results.3.Relying on the correspondence between the jammer and the radar established by the resource management link,this thesis regards jammers in formation as intelligent agents.To address the difficulty of the current cognitive jamming decision-making method based on reinforcement learning theory is difficult to meet the high real-time requirements for radar countermeasures,a cognitive jamming decision-making method against radar based on Asynchronous Advantage Actor-Critic(A3C)algorithm is proposed.We apply A3 C to the cognitive jamming decision-making field,design the overall framework,including the jammer model,the environment model,and the interaction mechanism,and develop a decision-making process.The jammer model uses an asynchronous multi-threaded way to interact with the environment model.The simulation results show that based on expanding the radar task transformation relationship table,compared with the cognitive jamming decision-making series methods based on Deep Q Network(DQN),this method can significantly improve time efficiency,decrease the average decision-making time by more than 30 times,and has obvious advantages in decision-making degree of accuracy.It shows that this method can provide powerful technical support for intelligent decision-making in multi-UAV cooperative radar jamming.