Research On Key Technology Of Damage Assessment For Unmanned Platform Cooperative Mission

With the rapid development of high-tech and the upgrading of weapons and equipment,unmanned platforms such as unmanned aerial vehicles,unmanned vehicles and unmanned submarines have been widely used in the battlefield,which can perform reconnaissance,surveillance and target attack in complex and dangerous environments.,electronic interference and other tasks have gradually become an indispensable important force in intelligent warfare.However,there are still two difficulties when the unmanned platform performs the damage assessment of cooperative tasks: 1)Affected by the complex battlefield environment and its own performance differences,the target damage data obtained by the unmanned platform is multi-modal,resulting in different unmanned aircraft.The platform’s damage assessment results for the same target are inconsistent.2)Before the unmanned platform performs the next strike mission on the target,due to many uncertain factors,it is difficult to determine the amount of ammunition required to achieve the desired damage effect.The main work of this paper is as follows:1)Aiming at the problem of inconsistent results when different unmanned platforms conduct damage assessment for multi-modal data,a blockchain-based unmanned platform collaborative damage assessment model is constructed.In this model,the nodes of the alliance blockchain are divided into 4 different types according to their responsibilities and functions,which are authentication supervision nodes,task release nodes,damage assessment nodes and information collection nodes.The authentication supervision nodes are responsible for the generation and distribution of group keys and the tracking of malicious node identities;the task publishing node publishes the task information of damage assessment to the alliance chain network;the information collection point is responsible for collecting the multi-modal damage data of the target;Modal damage data collaboratively evaluates the damage status of the target.Based on the cooperation and competition of the above four types of nodes,the model uses the unique smart contract mechanism of the blockchain to carry out a consensus on the consistency of the damage assessment results for the same target among the unmanned platforms,and combines the multi-modalities obtained by the unmanned platforms.The data processing of target damage status is a consistent conclusion.In addition,since the model uses a group signature mechanism,it ensures the authenticity of the data while also ensuring the security of the data.Simulation experiments verify the effectiveness of the model.2)Aiming at the problem that the ammunition consumption is difficult to determine when the unmanned platform performs the next strike mission on the target(ie the second strike problem),an improved QPSO-SVM ammunition consumption rapid prediction method is proposed.Firstly,an improved quantum particle swarm optimization algorithm is proposed,which adopts the nonlinear decreasing strategy to improve the scaling-expansion coefficient a of the quantum particle swarm in the optimization process,so as to reduce the possibility of falling into local optimum in the optimization process.Then,the optimization algorithm is used to obtain the nuclear parameters and penalty parameters,and the support vector machine is used to train the ammunition consumption model.Finally,the model is used to predict the ammunition consumption of the second strike.The experimental results show that the model has higher prediction accuracy and faster convergence speed when the training sample size is small and there are errors.Compared with other similar models,the accuracy and effectiveness of ammunition consumption prediction are There are obvious advantages in all aspects.