Vehicles Trajectory Prediction Using Recurrent VAE Network Articles uri icon

publication date

  • March 2022

start page

  • 32742

end page

  • 32749


  • 10

Electronic International Standard Serial Number (EISSN)

  • 2169-3536


  • This paper presents an analysis of the implementation and performance of a deep learning model based on Recurrent layers and Variational Auto Encoder model (VAE) architecture for prediction of future local trajectory and maneuver. The proposed method uses the encoder part of the VAE to represent the vehicle's surroundings agents behavior in time, taking advantage on the fact that VAE encodes similar situations or states close in the latent space and the generative properties of the VAE decoder, that is used to generate naturalistic driving trajectories. Furthermore, the variance of the predicted trajectory is estimated using the statistical properties of VAE model, increasing it if the input data is noisy or unrealistic and decreasing it if the model is certain about the prediction. The model is trained and evaluated with a public dataset. The results show that the proposed architecture outperforms state of the art methods in trajectory prediction error and provides a variance estimation that depends on input quality.


  • Mechanical Engineering


  • trajectory; predictive models; data models; roads; analytical models; deep learning; prediction algorithms learning (artificial intelligence); recurrent neural nets; trajectory control; vehicles trajectory prediction; recurrent vae network; deep learning model; recurrent layers; variational auto encoder model architecture; future local trajectory; maneuver; encoder part; vehicle; similar situations; generative properties; vae decoder; naturalistic driving trajectories; predicted trajectory; statistical properties; vae model; trajectory prediction error; intelligent vehicles; navigation; intelligent vehicles; land vehicles; prediction algorithms; trajectory prediction