Hierarchical Latent Structure for Multi-Modal Vehicle Trajectory Forecasting

TL;DR

This paper introduces a hierarchical latent structure in a VAE-based model to improve multi-modal vehicle trajectory forecasting, reducing ambiguity and increasing accuracy in predictions.

Contribution

It proposes a novel hierarchical latent VAE model that captures multi-modal trajectory distributions with lane and vehicle interaction context, outperforming existing methods.

Findings

Generates clear, multi-modal trajectory distributions.

Outperforms state-of-the-art models in prediction accuracy.

Effectively models lane and vehicle interactions.

Abstract

Variational autoencoder (VAE) has widely been utilized for modeling data distributions because it is theoretically elegant, easy to train, and has nice manifold representations. However, when applied to image reconstruction and synthesis tasks, VAE shows the limitation that the generated sample tends to be blurry. We observe that a similar problem, in which the generated trajectory is located between adjacent lanes, often arises in VAE-based trajectory forecasting models. To mitigate this problem, we introduce a hierarchical latent structure into the VAE-based forecasting model. Based on the assumption that the trajectory distribution can be approximated as a mixture of simple distributions (or modes), the low-level latent variable is employed to model each mode of the mixture and the high-level latent variable is employed to represent the weights for the modes. To model each mode accurately, we condition the low-level latent variable using two lane-level context vectors computed in novel ways, one corresponds to vehicle-lane interaction and the other to vehicle-vehicle interaction. The context vectors are also used to model the weights via the proposed mode selection network. To evaluate our forecasting model, we use two large-scale real-world datasets. Experimental results show that our model is not only capable of generating clear multi-modal trajectory distributions but also outperforms the state-of-the-art (SOTA) models in terms of prediction accuracy. Our code is available at https://github.com/d1024choi/HLSTrajForecast.