A Multi-Modal States based Vehicle Descriptor and Dilated Convolutional Social Pooling for Vehicle Trajectory Prediction

no code implementations • 7 Mar 2020 • Huimin Zhang, Yafei Wang, Junjia Liu, Chengwei Li, Taiyuan Ma, Chengliang Yin

Secondly, the LSTM encoder is used to encode the historical sequences composed of the vehicle descriptor and a novel dilated convolutional social pooling is proposed to improve modeling vehicles' spatial interactions.

Autonomous Vehicles Decision Making +1