CVMar 13, 2025

Trajectory Mamba: Efficient Attention-Mamba Forecasting Model Based on Selective SSM

arXiv:2503.10898v126 citationsh-index: 2CVPR
Originality Incremental advance
AI Analysis

This addresses computational bottlenecks in trajectory prediction for autonomous driving, offering a more efficient solution with incremental improvements in speed and parameter efficiency.

The paper tackles the problem of computational inefficiency in motion prediction for autonomous driving by introducing Trajectory Mamba, a framework based on selective state-space models that achieves linear time complexity. It demonstrates a four-fold reduction in FLOPs and over 40% fewer parameters while surpassing most previous methods on Argoverse datasets.

Motion prediction is crucial for autonomous driving, as it enables accurate forecasting of future vehicle trajectories based on historical inputs. This paper introduces Trajectory Mamba, a novel efficient trajectory prediction framework based on the selective state-space model (SSM). Conventional attention-based models face the challenge of computational costs that grow quadratically with the number of targets, hindering their application in highly dynamic environments. In response, we leverage the SSM to redesign the self-attention mechanism in the encoder-decoder architecture, thereby achieving linear time complexity. To address the potential reduction in prediction accuracy resulting from modifications to the attention mechanism, we propose a joint polyline encoding strategy to better capture the associations between static and dynamic contexts, ultimately enhancing prediction accuracy. Additionally, to balance prediction accuracy and inference speed, we adopted the decoder that differs entirely from the encoder. Through cross-state space attention, all target agents share the scene context, allowing the SSM to interact with the shared scene representation during decoding, thus inferring different trajectories over the next prediction steps. Our model achieves state-of-the-art results in terms of inference speed and parameter efficiency on both the Argoverse 1 and Argoverse 2 datasets. It demonstrates a four-fold reduction in FLOPs compared to existing methods and reduces parameter count by over 40% while surpassing the performance of the vast majority of previous methods. These findings validate the effectiveness of Trajectory Mamba in trajectory prediction tasks.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes