Lane Change Intention Prediction of two distinct Populations using a Transformer

2025-09-08

AI Summary

This study addresses the limited generalizability of lane-change intention prediction models by systematically evaluating the cross-regional transferability of Transformer architectures on driver behavior data from two distinct populations—Germany and Hong Kong. Existing approaches typically train models on single-population datasets and lack rigorous validation of cross-group robustness. We adopt a unified Transformer model that takes driver behavioral time-series data as input and conduct comparative experiments on real-world, vehicle-collected datasets from both regions. Results reveal that models trained exclusively on one region achieve only 39.43% accuracy when directly applied to the other region; in contrast, joint training on combined German and Hong Kong data boosts accuracy to 86.71%. This substantial improvement demonstrates the critical role of multi-population collaborative modeling in enhancing model generalizability. Our work establishes a new paradigm for developing transferable, robust driving intention prediction models and provides empirical evidence supporting cross-regional data integration in autonomous driving research.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Large Vision Models
• Application Domains: Mobility, Driving & Flight

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

As a result of the growing importance of lane change intention prediction for a safe and efficient driving experience in complex driving scenarios, researchers have in recent years started to train novel machine learning algorithms on available datasets with promising results. A shortcoming of this recent research effort, though, is that the vast majority of the proposed algorithms are trained on a single datasets. In doing so, researchers failed to test if their algorithm would be as effective if tested on a different dataset and, by extension, on a different population with respect to the one on which they were trained. In this article we test a transformer designed for lane change intention prediction on two datasets collected by LevelX in Germany and Hong Kong. We found that the transformer's accuracy plummeted when tested on a population different to the one it was trained on with accuracy values as low as 39.43%, but that when trained on both populations simultaneously it could achieve an accuracy as high as 86.71%. - This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.

Problem

• Testing lane change prediction models across different populations
• Addressing performance drop in cross-dataset transformer evaluation
• Improving algorithm generalization through multi-population training

Innovation

• Transformer model for lane change prediction
• Cross-dataset evaluation on German and Hong Kong populations
• Simultaneous training on multiple populations boosts accuracy