Dynamic allocation method for confrontation units

Abstract

The invention discloses a dynamic allocation method for confrontation units, and the method specifically comprises the steps: obtaining confrontation information in a digital battlefield environment,and preprocessing the confrontation information to obtain preprocessing information, using a trained recurrent neural network model for processing the grouping start vector as the initial state and the preprocessing information as the initial input to obtain first-step output, and using the attention model for processing the attribute information of the first-step output and the n adversarial units and the grouping end vector of the grouping end unit; sampling the normalization result according to a preset random sampling rule to obtain a sampling result; and repeating the above steps until the sampling result of this time is a component corresponding to the grouping ending unit. The grouping information obtained by the method is closer to the digital battlefield environment, is beneficialto completing confrontation tasks, and can adapt to the change of the situation of the digital battlefield.

Description

technical field [0001] The invention belongs to the technical field of digital battlefield simulation, and in particular relates to a dynamic allocation method for confrontation units. Background technique [0002] In the digital simulation battlefield environment, there are two opposing sides, and one of the opposing sides will launch a confrontation task against the other. In order to complete the confrontation task, the party will group its own units that can perform the confrontation task, generally divided into several groups, and members of the same group will perform the same or similar actions. In the following, the grouping is explained by taking the red team as one of the opposing parties and the blue team as the other opposing team. [0003] When the confrontation task is to penetrate the defense, the red team will often send several fighter jets to penetrate the defense at the same time; when the confrontation task is to approach and reconnaissance the blue team...

AI Technical Summary

Problems solved by technology

In order to simplify the control, the current grouping is generally a fixed group designated in advance. The fixed grouping has relatively large disadvantages. For example, some of them use a red unit as a group. During this mission, the attacking unit of the blue team may be shot down, and if only one red team is sent to perform this mission, it will still be shot down, resulting in the failure of the mission

In some cases, multiple red units form a group, and the number of red units in each group is the same. At this time, the members of the group when passing through the area with strong firepower of the blue side are the same as the members of the group when passing through the area with weak firepower of the blue side. cannot be done effectively

Generally speaking, in the current digital simulation battlefield, the allocation mode of confrontation units is relatively fixed, and corresponding strategies cannot be adopted according to changes in the external environment

Images