COTTA: Context-Aware Transfer Adaptation for Trajectory Prediction in Autonomous Driving
This work addresses the challenge of deploying autonomous driving models in new geographic regions, offering incremental improvements in transfer learning strategies for domain adaptation.
The paper tackled the problem of trajectory prediction models trained on Western data performing poorly in South Korean road environments due to domain discrepancies, and found that selectively fine-tuning the decoder while freezing the encoder reduced prediction error by over 66% compared to training from scratch.
Developing robust models to accurately predict the trajectories of surrounding agents is fundamental to autonomous driving safety. However, most public datasets, such as the Waymo Open Motion Dataset and Argoverse, are collected in Western road environments and do not reflect the unique traffic patterns, infrastructure, and driving behaviors of other regions, including South Korea. This domain discrepancy leads to performance degradation when state-of-the-art models trained on Western data are deployed in different geographic contexts. In this work, we investigate the adaptability of Query-Centric Trajectory Prediction (QCNet) when transferred from U.S.-based data to Korean road environments. Using a Korean autonomous driving dataset, we compare four training strategies: zero-shot transfer, training from scratch, full fine-tuning, and encoder freezing. Experimental results demonstrate that leveraging pretrained knowledge significantly improves prediction performance. Specifically, selectively fine-tuning the decoder while freezing the encoder yields the best trade-off between accuracy and training efficiency, reducing prediction error by over 66% compared to training from scratch. This study provides practical insights into effective transfer learning strategies for deploying trajectory prediction models in new geographic domains.