A Target Intention Recognition Method Based on Sequential Three-way Decision-Making

Abstract

The invention discloses a target intention recognition method based on sequential three-way decision-making, which introduces the sequential idea into the three-way decision-making, and divides the target intention recognition process into multiple stages according to the time sequence and the amount of target information obtained. In each stage, the multi-category three-way decision-making method is used to identify the target intention, and the delay decision is made for the situation where the target intention cannot be determined due to insufficient target information, and the target intention is continued to be re-identified by waiting for the next stage to obtain more target information , in order to solve the timing problem existing in the intent recognition, and improve the accuracy of target intent recognition.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and in particular relates to a target intention recognition method. Background technique [0002] In air combat, accurate and timely understanding of the enemy's combat intelligence can provide favorable support for our combat command and decision-making, and even change the direction of the war. If the combat intention of the enemy's target can be known in advance during the battle, the enemy's combat plan can be speculated, and then the target's combat movement and enemy's combat trend can be grasped. [0003] Intention recognition is a systematic problem. If the order of time and the incrementality of information acquisition are not considered, the timing characteristics and logical relationships in battlefield situation information cannot be effectively explained and utilized. Traditional intent recognition methods, such as template matching, knowledge graphs, and Bayesian reasoning, only ...

AI Technical Summary

Problems solved by technology

[0003] Intention recognition is a systematic problem. If the order of time and the incrementality of information acquisition are not considered, the timing characteristics and logical relationships in battlefield situation information cannot be effectively explained and utilized.

Traditional intent recognition methods, such as template matching, knowledge graphs, and Bayesian reasoning, only consider short-term or do not consider the changing characteristics of the battlefield situation, and the research on timing issues in target intent recognition is not deep enough. Timeliness issues should be considered, and at the same time, a comprehensive analysis should be made of the characteristics of using short-term and long-term battlefield situations

Images