测试大模型在真实驾驶中的表现，发现精度高不等于开得更好。
Closing the Loop: Motion Prediction Models beyond Open-Loop Benchmarks
- 对比顶级预测模型与规划器的协同效果，而非仅看离线精度。
- 参数减少86%的简化模型反而在闭环测试中表现更优。
- 强调时间一致性与规划兼容性比纯精度更重要，适合自动驾驶研发者。
受运动预测竞赛和基准测试推动，近年来出现了大量基于学习的预测模型，参数量达数百万，主要目标是将开放环路预测误差缩小至厘米级别。然而，这些基准无法评估模型集成到自动驾驶系统后的真实性能。本文系统评估了顶尖预测模型与运动规划器之间的相互作用，结果表明更高的开放环路精度并不总能带来更好的闭环驾驶行为；时间一致性与规划器兼容性等其他因素同样关键。此外，我们研究了这些模型的精简版本，令人意外的是，在某些情况下，参数量减少高达86%的模型在闭环驾驶表现上仍可媲美甚至优于原始模型。代码已开源：https://github.com/aumovio/pred2plan。
原文摘要 · Abstract (English)
Fueled by motion prediction competitions and benchmarks, recent years have seen the emergence of increasingly large learning based prediction models, many with millions of parameters, focused on improving open-loop prediction accuracy by mere centimeters. However, these benchmarks fail to assess whether such improvements translate to better performance when integrated into an autonomous driving stack. In this work, we systematically evaluate the interplay between state-of-the-art motion predictors and motion planners. Our results show that higher open-loop accuracy does not always correlate with better closed-loop driving behavior and that other factors, such as temporal consistency of predictions and planner compatibility, also play a critical role. Furthermore, we investigate downsized variants of these models, and, surprisingly, find that in some cases models with up to 86% fewer parameters yield comparable or even superior closed-loop driving performance. Our code is available at https://github.com/aumovio/pred2plan.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。