The ultimate goal of military intelligence is to equip the command and control (C2) system with the decision-making art of excellent human commanders and to be more agile and stable than human beings. Intelligent commander Alpha C2 solves the dynamic decision-making problem in the complex scenarios of air defense operations using a deep reinforcement learning framework. Unlike traditional C2 systems that rely on expert rules and decision-making models, Alpha C2 interacts with digital battlefields close to the real world and generates learning data. By integrating the states of multiple parties as input, a gated recurrent unit network is used to introduce historical information, and an attention mechanism selects the object of action, making the output decision more reliable. Without learning human combat experience, the neural network is trained in fixed-and random-strategy scenarios based on a proximal policy optimization algorithm. Finally, 1,000 rounds of offline confrontation were conducted on a digital battlefield, whose results show that the generalization ability of Alpha C2 trained using a random strategy is better, and that it can defeat an opponent with a higher winning rate than an Expert C2 system (72% vs 21%). The use of resources is more reasonable than Expert C2, reflecting the flexible and changeable art of command.

Deep Reinforcement Learning (DRL) methods are inefficient in the initial strategy exploration process in large-scale complex scenarios. This is becoming one of the bottlenecks in their application to large-scale game adversarial scenarios. This paper proposes a Safe reinforcement learning combined with Imitation learning for Task Assignment (SITA) method for a representative red-blue game confrontation scenario. Aiming at the problem of difficult sampling of Imitation Learning (IL), this paper combines human knowledge with adversarial rules to build a knowledge rule base; We propose the Imitation Learning with the Decoupled Network (ILDN) pre-training method to solve the problem of excessive initial invalid exploration; In order to reduce invalid exploration and improve the stability in the later stages of training, we incorporate Safe Reinforcement Learning (Safe RL) method after pre-training. Finally, we verified in the digital battlefield that the SITA method has higher training efficiency and strong generalization ability in large-scale complex scenarios.

Aiming at the problems of traditional dynamic weapon-target assignment algorithms in command decision-making, such as large computational amount, slow solution speed, and low calculation accuracy, combined with deep reinforcement learning theory, an improved Deep Deterministic Policy Gradient algorithm with dual noise and prioritized experience replay is proposed, which uses a double noise mechanism to expand the search range of the action, and introduces a priority experience playback mechanism to effectively achieve data utilization. Finally, the algorithm is simulated and validated on the ground-to-air countermeasures digital battlefield. The results of the experiment show that, under the framework of the deep neural network for intelligent weapon-target assignment proposed in this paper, compared to the traditional RELU algorithm, the agent trained with reinforcement learning algorithms, such as Deep Deterministic Policy Gradient algorithm, Asynchronous Advantage Actor-Critic algorithm, Deep Q Network algorithm performs better. It shows that the use of deep reinforcement learning algorithms to solve the weapon-target assignment problem in the field of air defense operations is scientific. In contrast to other reinforcement learning algorithms, the agent trained by the improved Deep Deterministic Policy Gradient algorithm has a higher win rate and reward in confrontation, and the use of weapon resources is more efficient. It shows that the model and algorithm have certain superiority and rationality. The results of this paper provide new ideas for solving the problem of weapon-target assignment in air defense combat command decisions.

With the rapid development of information technology, network-centric warfare (NCW) has become the main operational style now and even in the future, and the destruction resistance of networks has become one of the most important issues in the field. Based on the complex networks theory, the command and control (C2) network is constructed, and the topology characteristics of the network are analyzed. Aiming at the deliberate attack and random attack, the destruction resistance of C2 network is analyzed and optimized. The simulation results show that the C2 network conforms to the small-world and free-scale characteristics, and it is sensitive to deliberate attack. The strategy of low-degree edge addition can be adopted, that is, by expanding the horizontal contact between the same layer of combat entities and increasing the cross-layer contact between the different layers, the destruction resistance of C2 network can be effectively enhanced. The simulation results have a certain theoretical significance for the establishment of the C2 network and the optimization of destruction resistance of C2 network.

Spatial crowdsourcing is a mode that uses distributed artificial computing power to solve specific function sets through Internet outsourcing. It has broad application value in the networked command and control of current joint air defense operations. In this paper, we introduce the spatial crowdsourcing theory into the field of target allocation for joint air defense operations and establish a weapon-target assignment model based on spatial crowdsourcing mode, which is more appropriate to the real situation and highlights the system cooperation capability of joint air defense operations. To solve the model, we propose a heuristic variable weight nonlinear learning factor particle swarm optimization (VWNF-PSO). This algorithm can significantly improve the efficiency and adaptability to weapon-target assignment problems under large-scale extreme conditions. Finally, we establish two kinds of joint air defense operation scenarios to verify the proposed model, then compare the proposed algorithm with variable weight PSO (VWPSO) and adaptive learning factor PSO (AFPSO), to validate the effectiveness and efficiency of the VWNF-PSO algorithm proposed in this paper.

Aiming at the problem of multisensor resource scheduling in missile early warning operation, a scheduling decomposition strategy for missile early warning tasks under cooperative detection is proposed. Taking the detection benefit factor, target threat factor, and handover factor as the fitness function, we establish a sensor-subtask assignment (SSA) model and propose a hybrid discrete artificial bee colony (HDABC) algorithm to solve the optimal solution of the SSA model. The HDABC algorithm has the following improvements: in the initialization stage, a sensor-subtask-based coding method is designed to reduce the solution dimension, and the heuristic rules are used to obtain excellent populations to improve the convergence speed; in the employed bee and onlooker bee stage, a food source update strategy based on discrete differential mutation (DDM) operation is proposed to improve the searchability of the algorithm, and a sorting-based adaptive probability (SAP) selection method is applied to enhance the global search and local optimization capacities. Simulation experiments were carried out in operation scenarios of different scales. Experimental results showed that the proposed HDABC algorithm can obtain the optimal scheduling schemes and had a better solving performance when solving the SSA model, especially in the medium-scale and large-scale operation scenarios.