The limited deck space and highly dynamic environment pose significant challenges for autonomous dispatching of carrier-based aircraft. While existing reinforcement learning-based automatic parking techniques offer novel technological insights for autonomous carrier aircraft dispatching, these methods encounter non-convergence issues when directly applied to dynamic environments with constrained aircraft postures. To address this limitation, this paper proposes a locally guided reinforcement learning approach for carrier aircraft autonomous dispatching. The method introduces dual reward mechanisms: a reference trajectory-based local target state reward and a local state grid reward near the dispatching endpoint. These mechanisms effectively guide the learning process, preventing both local optima entrapment and convergence failure during training, thereby significantly enhancing the success rate of autonomous carrier aircraft dispatching. Experimental results demonstrate that the proposed approach outperforms conventional autonomous dispatching methods in terms of both success rate and operational safety. The method’s effectiveness has been validated in various mission scenarios and different carrier aircraft configurations.

Accurate recognition of flight deck operations for carrier-based aircraft, based on operation trajectories, is critical for optimizing carrier-based aircraft performance. This recognition involves understanding short-term and long-term spatial collaborative relationships among support agents and positions from long spatial–temporal trajectories. While the existing methods excel at recognizing collaborative behaviors from short trajectories, they often struggle with long spatial– temporal trajectories. To address this challenge, this paper introduces a dynamic graph method to enhance flight deck operation recognition. First, spatial–temporal collaborative relationships are modeled as a dynamic graph. Second, a discretized and compressed method is proposed to assign values to the states of this dynamic graph. To extract features that represent diverse collaborative relationships among agents and account for the duration of these relationships, a biased random walk is then conducted. Subsequently, the Swin Transformer is employed to comprehend spatial–temporal collaborative relationships, and a fully connected layer is applied to deck operation recognition. Finally, to address the scarcity of real datasets, a simulation pipeline is introduced to generate deck operations in virtual flight deck scenarios. Experimental results on the simulation dataset demonstrate the superior performance of the proposed method.