What question did this study set out to answer?

To resolve vanishing gradient limitations and simultaneously model spatial and temporal vehicle interactions for long-term trajectory prediction in dense traffic.

January 1, 2019IEEE AccessOpen Access

Modeling Vehicle Interactions via Modified LSTM Models for Trajectory Prediction

View Full Paper

Ask AI

Bookmark

Authors

SDShengzhe DaiUniversity Town of Shenzhen LLLi LiJiangsu University ZLZhiheng LiSouthwest Jiaotong University

Discussion

Member takes

Implication

Computational modeling study demonstrates improved vehicle trajectory prediction accuracy in dense highway traffic, indicating better handling of spatial interactions and long time series.

Key Points

To resolve vanishing gradient limitations and simultaneously model spatial and temporal vehicle interactions for long-term trajectory prediction in dense traffic.
Engineered a spatio-temporal long short-term memory (ST-LSTM) model embedding spatial interactions between neighboring vehicles.
Added shortcut connections across consecutive LSTM layers to mitigate vanishing gradient problems during long-sequence training.
Evaluated model performance against the baseline maneuver-LSTM (M-LSTM) architecture using the public I-80 and US-101 vehicle trajectory datasets.
The ST-LSTM model achieved higher trajectory prediction accuracy than the state-of-the-art maneuver-LSTM across both benchmark datasets.

Cite This Study

Dai et al. (2019) studied this question.

synapsesocial.com/papers/69dbe0c978a3e0e288685e47 https://doi.org/10.1109/access.2019.2907000

View Full Paper

Ask AI

Bookmark

Implication

Computational modeling study demonstrates improved vehicle trajectory prediction accuracy in dense highway traffic, indicating better handling of spatial interactions and long time series.

Modeling Vehicle Interactions via Modified LSTM Models for Trajectory Prediction

Discussion

Member takes

Implication

Key Points

Cite This Study

Modeling Vehicle Interactions via Modified LSTM Models for Trajectory Prediction

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider