Streaming Motion Forecasting for Autonomous Driving

TL;DR

This paper introduces a streaming forecasting benchmark for autonomous driving, addressing the limitations of snapshot-based methods by focusing on continuous data and occlusion handling, and proposes a plug-and-play algorithm to improve prediction accuracy and temporal coherence.

Contribution

The paper presents a new streaming forecasting benchmark and a versatile meta-algorithm that enhances existing snapshot-based forecasters for continuous, occlusion-aware trajectory prediction.

Findings

25% reduction in endpoint errors for occluded agents

10-20% reduction in trajectory fluctuations

Effective adaptation of snapshot forecasters to streaming data

Abstract

Trajectory forecasting is a widely-studied problem for autonomous navigation. However, existing benchmarks evaluate forecasting based on independent snapshots of trajectories, which are not representative of real-world applications that operate on a continuous stream of data. To bridge this gap, we introduce a benchmark that continuously queries future trajectories on streaming data and we refer to it as "streaming forecasting." Our benchmark inherently captures the disappearance and re-appearance of agents, presenting the emergent challenge of forecasting for occluded agents, which is a safety-critical problem yet overlooked by snapshot-based benchmarks. Moreover, forecasting in the context of continuous timestamps naturally asks for temporal coherence between predictions from adjacent timestamps. Based on this benchmark, we further provide solutions and analysis for streaming forecasting. We propose a plug-and-play meta-algorithm called "Predictive Streamer" that can adapt any snapshot-based forecaster into a streaming forecaster. Our algorithm estimates the states of occluded agents by propagating their positions with multi-modal trajectories, and leverages differentiable filters to ensure temporal consistency. Both occlusion reasoning and temporal coherence strategies significantly improve forecasting quality, resulting in 25% smaller endpoint errors for occluded agents and 10-20% smaller fluctuations of trajectories. Our work is intended to generate interest within the community by highlighting the importance of addressing motion forecasting in its intrinsic streaming setting. Code is available at https://github.com/ziqipang/StreamingForecasting.