An Enhanced Vehicle Trajectory Prediction Model Leveraging LSTM and Social-Attention Mechanisms

Senyao Qiao,Fei Gao,Jianghang Wu,Rui Zhao

DOI: https://doi.org/10.1109/access.2023.3345643

IF: 3.9

2023-01-01

IEEE Access

Abstract:Accurate trajectory prediction for multiple vehicles in complex social interaction environments is essential for ensuring the safety of autonomous vehicles and improving the quality of their planning and control. The social interactions between vehicles significantly influence their future trajectories. However, traditional trajectory prediction models based on Recurrent Neural Networks (RNN) or Convolutional Neural Networks (CNN) often overlook or simplify these interactions. Although these models may exhibit high performance in short-term predictions, they fail to achieve high prediction accuracy in scenarios with long-term dynamic interactions. To address this limitation, we propose a Social-Attention Long Short-Term Memory (LSTM) model which predicts the future trajectories of neighboring vehicles and achieves increased accuracy. Our proposed model employs a Social-Pooling layer to effectively capture cooperative behaviors and mutual influences between vehicles. Additionally, we incorporate a self-attention mechanism to weight the inputs and outputs of the Social-Pooling layer, which is significant for assessing the influence between vehicles in different positions. This combination allows our model to take into consideration both the dependencies within the sequence and the social relationships between vehicles, providing a more comprehensive scene understanding. The efficacy of our model is tested on two real-world freeway trajectory datasets, namely NGSIM and HighD. Our model surpasses various baseline methods, exhibiting exceptional accuracy in both prediction and tracking.

computer science, information systems,telecommunications,engineering, electrical & electronic

What problem does this paper attempt to address?

The paper attempts to address the problem of accurately predicting multiple vehicle trajectories in complex social interaction environments to ensure the safety of autonomous vehicles and improve the quality of their planning and control. Social interactions significantly influence the future trajectories of vehicles; however, traditional trajectory prediction models based on Recurrent Neural Networks (RNN) or Convolutional Neural Networks (CNN) often ignore or simplify these interactions. Although these models perform well in short-term predictions, they fail to achieve high-precision predictions in scenarios with long-term dynamic interactions.

To address this limitation, the authors propose a model that combines Long Short-Term Memory (LSTM) and Social-Attention mechanisms—Social-Attention Long Short-Term Memory model (Social-Attention LSTM, AS-LSTM). This model effectively captures the cooperative behaviors and mutual influences between vehicles through a Social-Pooling layer and introduces a self-attention mechanism to weight the inputs and outputs of the Social-Pooling layer, thereby assessing the influence between vehicles at different positions. This combination allows the model to consider dependencies within the sequence as well as social relationships between vehicles, providing a more comprehensive understanding of the scene. Experimental results show that the model outperforms various baseline methods on two real-world highway trajectory datasets (NGSIM and HighD), demonstrating excellent performance in prediction and tracking.