Research Article
[Retracted] Vehicle Motion State Prediction Method Integrating Point Cloud Time Series Multiview Features and Multitarget Interactive Information
Table 5
MSE statistics of the target vehicle global position Y of each prediction model in different interaction scenarios and based on point cloud multiview.
| Combination type of scenario and multiview | FC | NN | LSTM | BiLSTM |
| Single-vehicle operation | 0.8603 | 0.8059 | 0.0442 | 0.0324 | Single-vehicle operation + multiview | 0.8206 | 0.8041 | 0.0311 | 0.0262 | Two-vehicle interaction | 0.9565 | 0.8628 | 0.0753 | 0.0584 | Two-vehicle interaction + multiview | 0.7057 | 0.6706 | 0.0391 | 0.0301 | Three-vehicle interaction | 1.1467 | 0.9292 | 0.0674 | 0.0482 | Three-vehicle interaction + multiview | 0.8440 | 0.7958 | 0.0282 | 0.0134 | Four-vehicle interaction | 1.0361 | 0.8102 | 0.0729 | 0.0648 | Four-vehicle interaction + multiview | 0.9250 | 0.6871 | 0.0606 | 0.0182 |
|
|