Trajectory Prediction Method for Unmanned Systems Based on Dynamic Graph Neural Network with Adaptive Weight Adjustment

In response to the complexity and uncertainty of trajectory prediction for unmanned systems in highly dynamic and uncertain complex environments, this study proposes a dynamic graph neural network (D-GNN) method based on adjustment mechanism. This method consists of a weight adjustment network module and a dynamic graph of unmanned systems. The weight adjustment network module dynamically adjusts the weights, enhancing the model's adaptability to complex environmental diversity and dynamics, thereby effectively addressing the dynamic relationship modeling between nodes and edges. Experimental results show that this method performs well in trajectory prediction tasks, achieving high prediction accuracy. This method demonstrates higher efficiency and accuracy in handling tasks of different scales and complexities compared to the original dynamic graph neural network (D-GNN) without the adaptive weight adjustment module. Experimental results show that the proposed method outperforms the baseline model in terms of prediction accuracy and computational efficiency, particularly in complex and dynamic environments.