CILF:Causality Inspired Learning Framework for Out-of-Distribution Vehicle Trajectory Prediction

TL;DR

This paper introduces a causality-inspired learning framework for vehicle trajectory prediction that improves out-of-distribution generalization by explicitly modeling causal and non-causal features.

Contribution

The paper proposes the CILF framework with an Out-of-Distribution Causal Graph to better capture causal structures, enhancing domain generalization in trajectory prediction.

Findings

CILF outperforms baseline models on NGSIM and INTERACTION datasets.

Explicit causal modeling improves out-of-distribution robustness.

The framework achieves significant domain generalization improvements.

Abstract

Trajectory prediction is critical for autonomous driving vehicles. Most existing methods tend to model the correlation between history trajectory (input) and future trajectory (output). Since correlation is just a superficial description of reality, these methods rely heavily on the i.i.d. assumption and evince a heightened susceptibility to out-of-distribution data. To address this problem, we propose an Out-of- Distribution Causal Graph (OOD-CG), which explicitly defines the underlying causal structure of the data with three entangled latent features: 1) domain-invariant causal feature (IC), 2) domain-variant causal feature (VC), and 3) domain-variant non-causal feature (VN ). While these features are confounded by confounder (C) and domain selector (D). To leverage causal features for prediction, we propose a Causal Inspired Learning Framework (CILF), which includes three steps: 1) extracting domain-invariant causal feature by means of an invariance loss, 2) extracting domain variant feature by domain contrastive learning, and 3) separating domain-variant causal and non-causal feature by encouraging causal sufficiency. We evaluate the performance of CILF in different vehicle trajectory prediction models on the mainstream datasets NGSIM and INTERACTION. Experiments show promising improvements in CILF on domain generalization.