| ID | 原文 | 译文 |
| 3633 | 针对室内定位与定向系统，提出了一种结合粒子群优化（PSO）和天牛须搜索（BAS）的 VLC 室内定位与定向算法，通过 PSO 算法探索接收端的最优方向，利用 BAS 算法对每个粒子当前方向下的最佳三维坐标进行搜索。 | For the insufficient research on the simultaneously positioning and orientating system, a VLC positioning andorientating algorithm based on combining particle swarm optimization (PSO) and beetle antennae search (BAS) wasproposed. The PSO algorithm was used to explore the optimal direction of the receiver and meanwhile the BAS algorithmwas used to search the best three-dimensional coordinates of each particle in the current direction. |
| 3634 | 首先在 3 m×3 m×5 m 的室内空间和 40 dB 信噪比下进行仿真研究，所提出的 VLC 室内定位与定向系统实现了平均定位误差 4.82 cm 和平均定向误差 2.24°的性能指标。 | Firstly, the indoorspace of 3 m×3 m×5 m and the SNR of 40 dB were assumed in the simulation, the proposed VLC indoor positioning andorientating system could achieve the average positioning error of 4.82 cm and average orientating error of 2.24°. |
| 3635 | 然后在 0.9 m×0.9 m×1.5 m 的实验系统中，实施了同时定位与定向的实验验证，平均定位与定向误差分别为 5.32 cm 与 5.99°。 | Then, inthe laboratory space of 0.9 m×0.9 m×1.5 m, the experimental demonstration was accomplished for the first time, and theaverage positioning and orientating errors of this experimental system were 5.32 cm and 5.99°, respectively. |
| 3636 | 相比传统 VLC 室内定位方案，所提 VLC 室内定位与定向系统不需要接收端高度和方向的先验知识，大大降低了使用复杂度，具有更加广泛的应用场景。 | Compared with the traditional VLC indoor positioning schemes, the proposed VLC positioning and orientating system does not needthe prior knowledge of the height and direction of the receiver, which greatly reduces the system complexity and is ap-plicable to a wider range of applications. |
| 3637 | 在复杂的城市环境中，由于存在难以避免的 GNSS 定位信号中断现象以及车辆行驶过程中的误差累积，易造成所收集的车辆轨迹数据不准确和不完备，因此提出一种基于双向 RNN 的私家车轨迹重构算法， | To address the problem that in the complex urban environment, due to the inevitable interruption of GNSS po-sitioning signal and the accumulation of errors during vehicle driving, the collected vehicle trajectory data was likely tobe inaccurate and incomplete. a bidirectional weighted trajectory reconstruction algorithm was proposed based on RNNneural network. |
| 3638 | 使用了GNSS-OBD 轨迹采集设备收集车辆轨迹信息，利用多源数据融合实现双向加权轨迹重构。 | The GNSS-OBD trajectory acquisition device was used to collect vehicle trajectory information, andmulti-source data fusion was adopted to achieve bidirectional weighted trajectory reconstruction. |
| 3639 | 同时，在轨迹重构模型中引入神经算术逻辑单元（NALU），加强深度网络的外推能力并确保轨迹预测的精度，提高算法在应对城市复杂路段时轨迹重构的稳健性； | Furthermore, the neuralarithmetic logic unit (NALU) was leveraged with the purpose of enhancing the extrapolation ability of deep network andensuring the accuracy of trajectory reconstruction. |
| 3640 | 选取了实际城市路段进行了测试实验，并和现有算法进行了对比分析。 | For the evaluation, real-world experiments were conducted to evaluatethe performance of the proposed method in comparison with existing methods. |
| 3641 | 通过均方根误差（RMSE）以及 Google Earth 对重构轨迹进行可视化展示，实验结果验证了所提算法的有效性和可靠性。 | The root mean square error (RMSE) indi-cator shows the algorithm accuracy and the reconstructed trajectory is visually displayed through Google Earth. Experi-mental results validate the effectiveness and reliability of the proposed algorithm. |
| 3642 | 为解决泊位占有率的预测精度随步长增加而下降的问题，提出了一种基于注意力机制的泊位占有率预测模型。 | To solve the problem that the berth occupancy prediction accuracy decreases while the prediction step was in-creasing, a berth occupancy prediction model based on an attention mechanism was proposed, |