arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.08975cs.AI
收藏

通过高效轨迹生成分析和优化Alpamayo 1的延迟

Latency Analysis and Optimization of Alpamayo 1 via Efficient Trajectory Generation

Graduate School of Automobile and Mobility(汽车与移动研究生院);Kookmin University(韩国釜山大学);Department of Electrical and Computer Engineering(电子与计算机工程系);Seoul National University(首尔国立大学)

Yunseong Jeon, Namcheol Lee, Yoonsu Lee, Jangwoon Park, Sol Ahn, Jong-Chan Kim, Seongsoo Hong

AI总结:

本文通过将Alpamayo 1从多推理改为单推理系统,减少推理延迟并保持轨迹多样性,同时通过消除不必要的复制操作和低效内核执行提升扩散动作生成效率,实现了69.23%的延迟降低。

摘要:

基于推理的端到端自动驾驶(E2E AD)最近出现,因其能生成人类可读的推理并预测轨迹而备受关注。此类方法通常生成多个轨迹以捕捉多样化的未来行为,分为两种类型:(1)多推理,每个轨迹生成一个推理序列;(2)单推理,所有轨迹共享一个推理。前者在多样性方面更丰富,但计算冗余;后者更高效,但通常被认为牺牲了多样性。Alpamayo 1作为代表性系统,采用多推理方法并实现了具有竞争力的轨迹预测性能。然而,这种设计的效率仍鲜有研究,因此值得深入探讨。本文通过两种方式系统分析和改进Alpamayo 1。首先,通过将Alpamayo 1重设计为单推理系统,减少推理延迟同时保持轨迹多样性。通过大量实验发现,用单推理替代多推理不会显著降低轨迹多样性。其次,通过消除因不必要的复制操作和低效内核执行产生的块间开销,加速扩散动作生成。通过闭环和开环实验验证了这两种优化,证明在保持轨迹多样性和预测质量的同时,推理延迟减少了69.23%。这些结果突显了联合分析系统架构和运行时执行以提高基于推理的E2E AD系统效率的重要性。

英文摘要:

Reasoning-based end-to-end (E2E) autonomous driving has recently emerged as a promising approach to improving the interpretability of driving decisions as it can generate human-readable reasoning together with predicted trajectories. Such approaches commonly generate multiple trajectories to capture diverse future behaviors, and they fall into two categories: (1) multi-reasoning, where one reasoning sequence is generated per trajectory, and (2) single-reasoning, where a single reasoning is shared across all trajectories. The former offers richer diversity at the cost of redundant computation, while the latter is more efficient but is often assumed to sacrifice diversity. Alpamayo 1, a representative system, adopts the multi-reasoning approach and achieves competitive trajectory prediction performance. However, the efficiency of this design remains largely unexplored, making it a well-motivated subject for investigation. In this paper, we systematically analyze and improve Alpamayo 1 in two ways. First, we reduce inference latency while preserving trajectory diversity by redesigning Alpamayo 1 into a single-reasoning system. Through extensive experiments, we find that replacing multi-reasoning with single-reasoning does not meaningfully degrade trajectory diversity. Second, we accelerate diffusion-based action generation by eliminating inter-block overhead arising from unnecessary copy operations and inefficient kernel execution. Through closed-loop and open-loop experiments, we validate both optimizations, demonstrating a 69.23% reduction in inference latency while maintaining trajectory diversity and prediction quality. These results highlight the importance of jointly analyzing system architecture and runtime execution to improve the efficiency of reasoning-based E2E AD systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑