Skip to main content
To KTH's start page

Interaction-Aware Vehicle Trajectory Prediction via Attention Mechanism and Beyond

Time: Tue 2022-06-21 14.00 - 15.00

Location: Zoom link https://kth-se.zoom.us/j/68330181904

Video link: https://kth-se.zoom.us/j/68330181904

Language: English

Respondent: Wenxuan Wu , DCS

Opponent: Ning Wang

Supervisor: Yuchao Li

Examiner: Jonas Mårtensson

Export to calendar

Abstract: With the development of autonomous driving technology, vehicle trajectory prediction 
has become a hot topic in the intelligent traffic area. However, complex road conditions 
may bring multiple challenges to the vehicle trajectory prediction model. To address 
this, most recent studies mainly focus on designing different neural network structures 
to learn vehicles’ dynamics and interaction features for better prediction. In this 
thesis we restrict our research scope to highway scenarios. Based on the experimental 
comparison among Vanilla-RNN, Vanilla-LSTM, and Vanilla-Transformer, we find the 
best configuration of the dynamics-only encoder module and utilized it to design a 
novel model called the LSTM-attention model for vehicle trajectory prediction. The 
objective of our design is to explore whether the self-attention mechanism based 
encoder outperforms the pooling mechanism based encoder utilized in most current 
baseline models. The experiment results on the interaction encoder module show that 
the self-attention based encoder with 8 heads outperforms the pooling based encoder 
in longer prediction horizons. To test the robustness of our LSTM-attention model, 
we also compare the prediction performance between using maneuver-based decoder 
and using maneuver-free decoder, respectively. And we find the maneuver-based 
decoder performs better on the heavily unbalanced NGSIM dataset. Finally, to explore 
other latent interaction features our LSTM-attention model might fuse, we analyze the 
graph-based encoder and the polar-based encoder, respectively. Based on this, we find 
more meaningful designs we might exploit in our future work.