翻译 NGSIM数据评估结果表明：与LSTM编-解码器、1DCNN-LSTM和GRU编-解码器相比CGSTA-WG模型在长时域的轨迹预测上具有明显的优势能够使用较少的输入信息获得更准确的预测结果且更贴合不同类型的实际轨迹。

The NGSIM data evaluation results indicate that compared to the LSTM encoder-decoder, 1DCNN-LSTM, and GRU encoder-decoder models, the CGSTA-WG model has significant advantages in long-term trajectory prediction. It can obtain more accurate prediction results with less input information, and is more fitting for different types of actual trajectories.