Target maneuver behavior recognition method based on double drive model

Abstract

In order to solve the problem caused by the lack and imbalance of sample data and behavior knowledge of sample targets, an intelligent recognition method based on dual driving of data and knowledge is proposed. The idea of integrated learning is adopted to fuse the deep learning driven by data and the reasoning model driven by knowledge, which can not only make up for the poor performance of reasoning models in complex environments, but also increase the interpretability and robustness of deep learning. The effectiveness of the proposed method is verified by simulation and comparison experiments. The results show that the proposed algorithm is superior to the single driven system based only on knowledge driven or data driven, with an accuracy rate of 90.1%. This method greatly improves the adaptability to complex environments and recognition accuracy, and has important engineering application value for target maneuver behavior recognition.

Keywords

• maneuver recognition|situational cognition|data driven|knowledge driven