Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

X-TRACK: Physics-Aware xLSTM for Realistic Vehicle Trajectory Prediction

Aanchal Rajesh Chugh, Marion Neumeier, Sebastian Dorn

TL;DR

This work presents X-TRAJ, an xLSTM-based highway trajectory predictor, and its physics-aware extension X-TRACK, which integrates a non-holonomic kinematic layer to enforce physically feasible motion. The architecture combines an xLSTM encoder, a graph attention network for social interactions, and a decoder, with X-TRACK adding a kinematic layer that outputs motion parameters $(a_x,\dot{\psi})$ rather than direct positions. On highD and NGSIM, X-TRACK achieves state-of-the-art ADE/FDE and substantial RMSE gains, particularly at short horizons, demonstrating that physics priors can markedly improve both accuracy and realism. The results highlight the value of hybrid data-driven and physics-guided models for safe, reliable autonomous driving trajectory prediction, and point to future work incorporating richer map information and urban driving scenarios.

Abstract

Recent advancements in Recurrent Neural Network (RNN) architectures, particularly the Extended Long Short Term Memory (xLSTM), have addressed the limitations of traditional Long Short Term Memory (LSTM) networks by introducing exponential gating and enhanced memory structures. These improvements make xLSTM suitable for time-series prediction tasks as they exhibit the ability to model long-term temporal dependencies better than LSTMs. Despite their potential, these xLSTM-based models remain largely unexplored in the context of vehicle trajectory prediction. Therefore, this paper introduces a novel xLSTM-based vehicle trajectory prediction framework, X-TRAJ, and its physics-aware variant, X-TRACK (eXtended LSTM for TRAjectory prediction Constraint by Kinematics), which explicitly integrates vehicle motion kinematics into the model learning process. By introducing physical constraints, the proposed model generates realistic and feasible trajectories. A comprehensive evaluation on the highD and NGSIM datasets demonstrates that X-TRACK outperforms state-of-the-art baselines.

X-TRACK: Physics-Aware xLSTM for Realistic Vehicle Trajectory Prediction

TL;DR

This work presents X-TRAJ, an xLSTM-based highway trajectory predictor, and its physics-aware extension X-TRACK, which integrates a non-holonomic kinematic layer to enforce physically feasible motion. The architecture combines an xLSTM encoder, a graph attention network for social interactions, and a decoder, with X-TRACK adding a kinematic layer that outputs motion parameters rather than direct positions. On highD and NGSIM, X-TRACK achieves state-of-the-art ADE/FDE and substantial RMSE gains, particularly at short horizons, demonstrating that physics priors can markedly improve both accuracy and realism. The results highlight the value of hybrid data-driven and physics-guided models for safe, reliable autonomous driving trajectory prediction, and point to future work incorporating richer map information and urban driving scenarios.

Abstract

Recent advancements in Recurrent Neural Network (RNN) architectures, particularly the Extended Long Short Term Memory (xLSTM), have addressed the limitations of traditional Long Short Term Memory (LSTM) networks by introducing exponential gating and enhanced memory structures. These improvements make xLSTM suitable for time-series prediction tasks as they exhibit the ability to model long-term temporal dependencies better than LSTMs. Despite their potential, these xLSTM-based models remain largely unexplored in the context of vehicle trajectory prediction. Therefore, this paper introduces a novel xLSTM-based vehicle trajectory prediction framework, X-TRAJ, and its physics-aware variant, X-TRACK (eXtended LSTM for TRAjectory prediction Constraint by Kinematics), which explicitly integrates vehicle motion kinematics into the model learning process. By introducing physical constraints, the proposed model generates realistic and feasible trajectories. A comprehensive evaluation on the highD and NGSIM datasets demonstrates that X-TRACK outperforms state-of-the-art baselines.

Paper Structure

This paper contains 20 sections, 11 equations, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Deep Learning-Based Trajectory Prediction
  4. Physics-Based Trajectory Prediction
  5. xLSTM-based Vehicle Trajectory Prediction
  6. Preliminaries
  7. Problem Definition
  8. Overall Framework
  9. Historical Encoder
  10. Modeling Vehicle Interactions
  11. Future Decoder
  12. Physics-based Kinematic Layer
  13. Dataset, Experiments and Evaluation
  14. Datasets
  15. Training and Implementation Details
  16. ...and 5 more sections

Figures (2)

  • Figure 1: Proposed X-TRACK architecture: The sLSTM block generates an encoded vector for all the vehicles in the traffic scenario. The GAT layers model the vehicle interactions between the target vehicle (shown in red) and the neighboring vehicles based on attention scores. The concatenation of the output from the GAT module and target vehicle encoding is passed through the decoder to predict future motion parameters. The Kinematic layer then transforms these motion parameters into position coordinates to provide the future trajectory of the target vehicle.
  • Figure 2: Comparison of predicted trajectories on the highD dataset using X-TRAJ and X-TRACK in case of (a) Keep lane (b) Lane change.