In this paper, a cooperative target assignment algorithm for multiple missiles against unmanned aerial vehicle (UAV) swarms is proposed. First, each missile will intercept each UAV by simulation with proportional guidance law to obtain the final speed gap (FSG) when the missile and the UAV encounter. Then, taking the maximum sum of FSG of the assignment scheme as the optimization goal, and the giving maximum assigned missiles for a single target (MAM) as constraints, the problem is transformed to a non-equilibrium assignment problem with constraints, which can be solved by 0-1 linear programming. Meanwhile, a trajectory intersection judgement will be carried out for the optimum assignment scheme, and once trajectory intersection exists, one of the trajectories will be punished to avoid trajectory intersection, and the non-equilibrium assignment model will be modified and re-optimized until the optimum assignment scheme without trajectory intersection is obtained. Simulation results show that the proposed method can obtain the optimum assignment scheme for multiple missiles against UAV swarms in a short time, which can maximize the interception effects efficiently.

Timely and accurate system processing ability of missile electronic system against high-speed weak and small targets is one of the key factors for the missile to lock and successful strike the target. The computing resources are the core resources of target information processing. A hybrid architecture of target information intelligent processing based on heterogeneous computing resources of GPU and FPGA is proposed, which supports neural network algorithm based target information processing and software development by jointly utilizing GPU parallel processing and AI coprocessing ability and FPGA programmable extension ability. The standardization and generality design of proposed architecture meets the needs of on-demand customization, functional expansion and flexible configuration. The test results show that the proposed system can meet the requirements of in-time processing and agility of missile electronic system.

Aiming at the problems of large randomness and low search efficiency of rapid exploring random tree (RRT) path planning algorithm, combined with the characteristics that reinforcement learning can select strategies according to prior knowledge, an improved RRT optimization algorithm based on deep Q-network (DQN) is proposed. Firstly, the obstacle avoidance strategy with variable step in complex domain is designed, and the Markov decision process (MDP) model of random tree growth in RRT algorithm is established. Then, the obstacle avoidance strategy and MDP model are connected to the interface of RRT-Connect algorithm, and the specific process of training and path planning is designed. Finally, the simulation experiment is carried out on the MATLAB software platform. The simulation results show that the improved RRT-Connect algorithm based on deep Q-network (DQN-RRT-C) has a significant improvement in rapidity and search efficiency.

In the scenario of air-to-air missiles intercepting large-scale swarms, because the performance of centralized algorithms to solve large-scale task assignment problems is limited by communication capabilities and real-time performance, it is practically important to develop distributed algorithms to solve task assignment problems. Based on the hedonic coalition game (HCG) model, this paper designs a distributed missile swarm task assignment model, and uses the spatial adaptive play (SAP) algorithm in the game learning algorithm to achieve the optimal allocation solution. The simulation comparison experiment with the centralized algorithm shows that the distributed task assignment model based on the HCG designed in this paper can solve the distributed task allocation problem with constraints, and has comparable optimization performance and better convergence efficiency.

Landing guidance for reusable launch vehicle should ensure the accuracy of landing position and velocity as well as minimized fuel consumption. Landing guidance methods based on optimal control is based on accurate rocket dynamic model, which corrupts the scalability of guidance methods. To address this problem, a neural network landing guidance policy is developed based on model-free iterative reinforcement learning approach. First, a Markov decision process model of the rocket landing guidance problem is established, and a staged reward function is designed according to the terminal constraints and fuel consumption index; Further, a multilayer perceptron guidance policy network is developed, and a model-free proximal policy optimization algorithm is adopted to achieve iterative optimization of the guidance policy network through interaction with the rocket landing guidance Markov decision process; Finally, the guidance policy is validated under simulations of a reusable launch vehicle landing scenario. The results show that the proposed reinforcement learning landing guidance policy can achieve high landing accuracy, near optimal fuel consumption, and adaptivity to parameter uncertainty of the rocket model.

Low signature target group detection is a difficult problem. With the development of sensor technology, the target detection method based on multimodal data fusion becomes possible. Traditional methods focus on the artificial design of multi-modal data fusion level, by using signal level and decision-making level fusion, and neglecting multi-modal inherent fusion features. In order to make full use of the inherent multimodal feature expression ability of deep convolution network, a point to point multimodal data learning fusion network is proposed, which can fuse data from visible, infrared and Doppler pulse radar data devices. The experimental results on FLIR_ADAS dataset show that the proposed algorithm can significantly improve the performance of low signature target group detection.

The US military’s "Mosaic Warfare" combat concept from the perspectives of military strategy, combat theory, and battlefield application are summarized in this article. It has the characteristics of clear directionality, distinct inheritance, and outstanding innovation. It concludes that its combat concept and concept may be important to our combat system. Target protection, the seizure of electromagnetic power, and the suddenness of war have had an impact. On this basis, countermeasures and suggestions have been put forward in terms of combat theory, power building, equipment development, and application of tactics to deal with the challenges of the US military’s "Mosaic Warfare".

Aiming at the combat scenario of multiple air-to-air missiles attacking multiple aircrafts, the multi-objective task assignment of cluster collaboration is studied to intercept multiple aircrafts. In view of the highly dynamic and complex characteristics of the battlefield environment, a real-time task assignment scheme based on Grey Wolf Optimization (GWO) is adopted. In order to solve the shortcoming of the Grey Wolf Optimization, which is easy to fall into local optimum and easy to mature, two improvement strategies are proposed. One is to optimize the population, which uses the good point set theory to enhance the ergodicity of the population. The second is to improve the algorithm, using tabu search to update the position of the head to improve the global search ability of the algorithm, so that it can jump out of the local optimal region more easily, and adopt the adaptive adjustment strategy to speed up the convergence speed of the algorithm. Finally, the simulation experiment of multi-objective task assignment is carried out for a missile cluster consisting of two groups of heterogeneous missiles. The experimental results verify the feasibility and superiority of the algorithm, and meet the requirements of real-time dynamic task assignment.

In the future, the combat environment will be unmanned, intelligent and systematic. The background of the concept of Mosaic War is introduced; the relationships with the theory of OODA loop and the idea of decision centric warfare are pointed out; the core idea of Mosaic War and its application at the operational level and system construction level are analyzed. The key technologies are summarized to support the intelligent operation of mosaic warfare troops in terms of network information share, command and control of communication, as well as intelligent and hybrid decision making. Considering the advantages of Mosaic War in concealment and flexibility, and the deficiencies in communication, command & control and logistics, in order to delay and block the OODA fast closed loop, the countermeasures are proposed from the aspects of electromagnetic spectrum confrontation, intelligent algorithm and group confrontation.

Using AUTODYN software to carry out numerical simulation, the “up-down” method was used to calculate the initiation velocity threshold of composition B shielded by 6 mm/9 mm thick 45 steel cover plate impacted by 10 g tungsten alloy spherical fragment. According to the Richard M. Lloyd and Jacobs-Roslund model equations, engineering calculations under the same target conditions were carried out. The results show that the velocity threshold of 10 g tungsten alloy spherical fragment to detonate 6 mm/9 mm thick 45 steel shielded composition B is 2 100 m/s and 2 200 m/s respectively, and the calculated results based on Jacobs-Roslund model equation are more consistent with the numerical simulation results. The conclusion of this paper can provide a basis for the lethality design and assessment of anti-missile warhead.

An efficient method based on sequential surrogate-based optimization is proposed to realize the integrated design optimization for aerodynamic shape and trajectory of gliding vehicle. In order to obtain the aerodynamic characteristics of gliding vehicle, a parametric modeling method of the aerodynamic shape is developed, then the surface panel method based on unstructured triangular elements and streamline tracing technology are carried out to compute the viscous aerodynamic coefficient. The maximum range for given aerodynamic shape scheme is evaluated based on the adaptive Legendre-Gauss-Radau pseudospectral method. To improve the optimization efficiency, a multi-sampling efficient global optimization method based on Kriging surrogate model is proposed and numerically verified. The results show that this method demonstrates a high efficiency on implementing the integrated design optimization for aerodynamic shape and trajectory of gliding vehicle, and can significantly reduce the computational times of the time-consuming disciplinary analysis compared with the traditional method.

A sliding mode control law based on disturbance observer is designed to reduce the uncertainty and external disturbance in the longitudinal model of high Mach number aircraft and improve the convergence speed. First, the longitudinal model of high Mach number aircraft is given, and the state space form is obtained through model transformation. Then, an improved approach law is proposed for the transformed model, while a sliding mode control law based on interference observer is designed, and the stability of the closed-loop system is strictly proved by Lyapunov theorem. Finally, simulation results show that the sliding mode controller based on the disturbance observer is effective and has good robustness.

This paper studies the fusion estimation of multi-rate state distribution for multi-sensor network system with communication bandwidth constraints. For multi-rate sampled data system, a new expansion method is proposed, which transforms multi-rate sampled data system into single rate system. In order to meet the limited communication bandwidth, a dimension reduction strategy is used to allow only part of the local estimated signal components to be sent to the fusion center (FC). At the same time, a compensation strategy is proposed to compensate the unsent components of each component. For the local state estimation generated by multi-rate sensors, a new fusion estimation method is proposed by using the sequential fast covariance intersection (SFCI) method. Finally, a simulation example is provided to verify the effectiveness of the proposed distributed fusion estimation method.

As a new frontier of human activity, space is threatened by asteroid, space debris and military contest, and space security is closely related to the common interests of all mankind. Focusing on the topic of space security situation, a new definition of space security is proposed in this paper from the perspective of space domain and application scenario. In addition, considering both man-made threats and natural threats, current situation and near future development of space security is analyzed. And then, the main challenges of space security situation are discussed, such as asteroid defense, debris removal and outer space arms control. Lastly, four proposals about the big issue of global space security improvement are presented, from the point of view of diplomacy, law, technology, culture.

With the constant change of the concept of combat and the role of technology promotion, the rapid development and extensive application of new and high technologies with aerospace technology, new concept weapons, artificial intelligence, and information network as the core in the military field have brought about historic changes in the form of air combat, the mode of combat and the confrontation environment.With the continuous improvement of adversary air combat capability, the continuous change of combat environment and the development of new aircraft platform, higher requirements are put forward for the development of new airborne weapons. Therefore, starting from the air combat development system, combined with the concept of air combat and equipment development trends of the United States and other military powers, this paper combs the characteristics of the future air combat confrontation environment and its combat mode, and makes research judgment and suggestions on the future air combat concept, so as to provide reference for the future air combat mode research and new airborne weapon research in China.

The generation mechanism of U.S. weapon equipment operational requirements is introduced; the existing problem of PLA is analyzed by comparison in this area. Using the thought of system engineering, the generation mechanism of weapon equipment operational requirements is discussed, and the enlightenment and suggestions are putforward.