Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Digital-Twin Losses for Lane-Compliant Trajectory Prediction at Urban Intersections

Kuo-Yi Chao, Erik Leo Haß, Melina Gegg, Jiajie Zhang, Ralph Raßhofer, Alois Christian Knoll

TL;DR

Experimental results demonstrate that the proposed training scheme significantly reduces critical violations while maintaining comparable prediction accuracy and real-time performance, highlighting the potential of digital twin-driven multi-loss learning for V2X-enabled intelligent transportation systems.

Abstract

Accurate and safety-conscious trajectory prediction is a key technology for intelligent transportation systems, especially in V2X-enabled urban environments with complex multi-agent interactions. In this paper, we created a digital twin-driven V2X trajectory prediction pipeline that jointly leverages cooperative perception from vehicles and infrastructure to forecast multi-agent motion at signalized intersections. The proposed model combines a Bi-LSTM-based generator with a structured training objective consisting of a standard mean squared error (MSE) loss and a novel twin loss. The twin loss encodes infrastructure constraints, collision avoidance, diversity across predicted modes, and rule-based priors derived from the digital twin. While the MSE term ensures point-wise accuracy, the twin loss penalizes traffic rule violations, predicted collisions, and mode collapse, guiding the model toward scene-consistent and safety-compliant predictions. We train and evaluate our approach on real-world V2X data sent from the intersection to the vehicle and collected in urban corridors. In addition to standard trajectory metrics (ADE, FDE), we introduce ITS-relevant safety indicators, including infrastructure and rule violation rates. Experimental results demonstrate that the proposed training scheme significantly reduces critical violations while maintaining comparable prediction accuracy and real-time performance, highlighting the potential of digital twin-driven multi-loss learning for V2X-enabled intelligent transportation systems.

Digital-Twin Losses for Lane-Compliant Trajectory Prediction at Urban Intersections

TL;DR

Experimental results demonstrate that the proposed training scheme significantly reduces critical violations while maintaining comparable prediction accuracy and real-time performance, highlighting the potential of digital twin-driven multi-loss learning for V2X-enabled intelligent transportation systems.

Abstract

Accurate and safety-conscious trajectory prediction is a key technology for intelligent transportation systems, especially in V2X-enabled urban environments with complex multi-agent interactions. In this paper, we created a digital twin-driven V2X trajectory prediction pipeline that jointly leverages cooperative perception from vehicles and infrastructure to forecast multi-agent motion at signalized intersections. The proposed model combines a Bi-LSTM-based generator with a structured training objective consisting of a standard mean squared error (MSE) loss and a novel twin loss. The twin loss encodes infrastructure constraints, collision avoidance, diversity across predicted modes, and rule-based priors derived from the digital twin. While the MSE term ensures point-wise accuracy, the twin loss penalizes traffic rule violations, predicted collisions, and mode collapse, guiding the model toward scene-consistent and safety-compliant predictions. We train and evaluate our approach on real-world V2X data sent from the intersection to the vehicle and collected in urban corridors. In addition to standard trajectory metrics (ADE, FDE), we introduce ITS-relevant safety indicators, including infrastructure and rule violation rates. Experimental results demonstrate that the proposed training scheme significantly reduces critical violations while maintaining comparable prediction accuracy and real-time performance, highlighting the potential of digital twin-driven multi-loss learning for V2X-enabled intelligent transportation systems.
Paper Structure (35 sections, 11 equations, 2 figures, 4 tables)

This paper contains 35 sections, 11 equations, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Classical Motion Models
  4. Recurrent Trajectory Models
  5. Map-Aware Trajectory Prediction
  6. Auxiliary Loss Design
  7. Digital Twins for Autonomous Driving
  8. Dataset
  9. Recording and Parsing
  10. Trajectory Extraction
  11. Anchor-Relative Normalisation
  12. Feature Vector
  13. Methodology
  14. Model Architecture
  15. Training Losses
  16. ...and 20 more sections

Figures (2)

  • Figure 1: TUM intersection scene: OpenDRIVE lane map (grey, 2,233 segments) overlaid with recorded agent trajectories (946 objects, coloured by class) from the first 30,000 frames. The lane geometry serves as the infrastructure prior for the proposed twin-consistency loss.
  • Figure 2: LSTM encoder-decoder architecture. The dashed orange region highlights where MC-Dropout is active during inference for uncertainty estimation.