Adaptive Prediction Ensemble: Improving Out-of-Distribution Generalization of Motion Forecasting

TL;DR

The paper introduces Adaptive Prediction Ensemble (APE), a hybrid framework combining deep learning and rule-based models with a learned routing function, significantly enhancing out-of-distribution generalization in motion forecasting for autonomous driving.

Contribution

It proposes a novel adaptive ensemble method with a learned routing mechanism to improve OOD generalization in trajectory prediction models.

Findings

APE outperforms individual models on large-scale datasets.

The method improves zero-shot generalization across datasets.

Enhanced performance in long-horizon and high OOD scenarios.

Abstract

Deep learning-based trajectory prediction models for autonomous driving often struggle with generalization to out-of-distribution (OOD) scenarios, sometimes performing worse than simple rule-based models. To address this limitation, we propose a novel framework, Adaptive Prediction Ensemble (APE), which integrates deep learning and rule-based prediction experts. A learned routing function, trained concurrently with the deep learning model, dynamically selects the most reliable prediction based on the input scenario. Our experiments on large-scale datasets, including Waymo Open Motion Dataset (WOMD) and Argoverse, demonstrate improvement in zero-shot generalization across datasets. We show that our method outperforms individual prediction models and other variants, particularly in long-horizon prediction and scenarios with a high proportion of OOD data. This work highlights the potential of hybrid approaches for robust and generalizable motion prediction in autonomous driving. More details can be found on the project page: https://sites.google.com/view/ape-generalization.