融合图卷积网络与社交池化的多模态轨迹预测模型
赵庆慧,崔鑫,张艺炜,陈燕

Multimodal trajectory prediction model integrating graph convolutional networks and social pooling
Qinghui ZHAO,Xin CUI,Yiwei ZHANG,Yan CHEN
表 2 不同轨迹预测模型在HighD数据集上的性能对比
Tab.2 Performance comparison of different trajectory prediction models on dataset HighD
模型RMSE/mNLL
tp=1 stp=2 stp=3 stp=4 stp=5 s平均值tp=1 stp=2 stp=3 stp=4 stp=5 s平均值
CV0.330.781.622.433.671.7661.943.094.856.127.034.606
PiP0.240.681.342.193.421.5740.452.273.344.204.763.004
WSiP0.220.591.212.053.041.4220.341.832.723.494.252.526
HAN0.180.441.051.722.871.2520.271.242.373.063.952.178
HLTP++0.190.421.111.642.751.2220.291.382.593.043.872.234
DRBP0.180.491.111.622.721.2240.241.272.442.933.692.114
GC-LSTM0.150.370.991.542.651.1400.231.072.042.743.531.922