Forecasting the Past: Gradient-Based Distribution Shift Detection in Trajectory Prediction

2026-04-14

AI Summary

This work addresses the significant performance degradation of autonomous driving trajectory prediction models under test-time distribution shifts, which often leads to erroneous predictions in unfamiliar scenarios. The authors propose a self-supervised post-processing method that requires no modification to the original model: a decoder is trained to predict the latter half of a trajectory from its first half, and the L2 norm of the gradient of the prediction loss with respect to the decoder’s final-layer parameters serves as a distribution shift detection score. This approach efficiently identifies out-of-distribution inputs and enables early collision warnings. Experiments demonstrate that the method substantially outperforms existing approaches on the Shifts and Argoverse datasets and has been successfully integrated into a Deep Q-Network motion planner within the Highway simulation environment, achieving reliable collision risk detection.

Technology Category

Planning, Routing, and Scheduling: Model-Based ReasoningSearch and Optimization: Learning to SearchMachine Learning: Calibration & Uncertainty Quantification

Application Category

Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingSemantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsResponsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Trajectory prediction models often fail in real-world automated driving due to distributional shifts between training and test conditions. Such distributional shifts, whether behavioural or environmental, pose a critical risk by causing the model to make incorrect forecasts in unfamiliar situations. We propose a self-supervised method that trains a decoder in a post-hoc fashion on the self-supervised task of forecasting the second half of observed trajectories from the first half. The L2 norm of the gradient of this forecasting loss with respect to the decoder's final layer defines a score to identify distribution shifts. Our approach, first, does not affect the trajectory prediction model, ensuring no interference with original prediction performance and second, demonstrates substantial improvements on distribution shift detection for trajectory prediction on the Shifts and Argoverse datasets. Moreover, we show that this method can also be used to early detect collisions of a deep Q-Network motion planner in the Highway simulator. Source code is available at https://github.com/Michedev/forecasting-the-past.

Problem

• distribution shift
• trajectory prediction
• autonomous driving
• model robustness

Innovation

• distribution shift detection
• trajectory prediction
• self-supervised learning
• gradient-based scoring
• post-hoc adaptation