基于时空特征与交互关系建模的换道意图预测
高帅帅,魏诚,惠飞,张竟成,宋汉辰

Lane-changing intention prediction based on spatiotemporal feature and interaction modeling
Shuaishuai GAO,Cheng WEI,Fei HUI,Jingcheng ZHANG,Hanchen SONG
表 2 输入长度为1 s时不同模型的性能对比
Tab.2 Performance comparison of different models with 1 s input length
模型类别$ P $/%$ R $/%F1 /%$ {A}_{\text{cc}} $/%
Social-LSTM
(模型A)
LK81.2485.3783.2587.99
LLC90.6791.2490.96
RLC92.6687.3689.93
ST-GCN
(模型B)
LK86.8390.5688.6692.17
LLC94.3693.0393.69
RLC95.6892.9294.28
Transformer
(模型C)
LK94.3290.5192.3795.02
LLC95.4498.6997.04
RLC95.2695.8695.56
GATv2
(模型D)
LK70.2171.9571.0778.74
LLC82.4683.8083.12
RLC83.9380.4982.17
Transformer-GATv2
(模型E)
LK95.5393.0894.2996.24
LLC97.1598.4897.81
RLC96.0197.1796.59