SparScene: Efficient Traffic Scene Representation via Sparse Graph Learning for Large-Scale Trajectory Generation

TL;DR

SparScene introduces a sparse graph learning framework that efficiently models large-scale traffic scenes for autonomous driving, significantly improving scalability and inference speed while maintaining competitive trajectory prediction performance.

Contribution

The paper presents a novel sparse graph learning approach that leverages lane topology for efficient traffic scene representation, enabling scalable and fast multi-agent trajectory generation.

Findings

Generates trajectories for over 200 agents within 5 ms.

Scales to 5,000 agents and 17,000 lanes with 54 ms inference time.

Uses minimal GPU memory of 2.9 GB, demonstrating high efficiency.

Abstract

Multi-agent trajectory generation is a core problem for autonomous driving and intelligent transportation systems. However, efficiently modeling the dynamic interactions between numerous road users and infrastructures in complex scenes remains an open problem. Existing methods typically employ distance-based or fully connected dense graph structures to capture interaction information, which not only introduces a large number of redundant edges but also requires complex and heavily parameterized networks for encoding, thereby resulting in low training and inference efficiency, limiting scalability to large and complex traffic scenes. To overcome the limitations of existing methods, we propose SparScene, a sparse graph learning framework designed for efficient and scalable traffic scene representation. Instead of relying on distance thresholds, SparScene leverages the lane graph topology to construct structure-aware sparse connections between agents and lanes, enabling efficient yet informative scene graph representation. SparScene adopts a lightweight graph encoder that efficiently aggregates agent-map and agent-agent interactions, yielding compact scene representations with substantially improved efficiency and scalability. On the motion prediction benchmark of the Waymo Open Motion Dataset (WOMD), SparScene achieves competitive performance with remarkable efficiency. It generates trajectories for more than 200 agents in a scene within 5 ms and scales to more than 5,000 agents and 17,000 lanes with merely 54 ms of inference time with a GPU memory of 2.9 GB, highlighting its superior scalability for large-scale traffic scenes.