Abstract
Emergent-scene safety is the key milestone for fully autonomous driving, and reliable on-time prediction is essential to maintain safety in emergency scenarios. However, these emergency scenarios are long-tailed and hard to collect, which restricts the system from getting reliable predictions. In this paper, we build a new dataset, which aims at the long-term prediction with the inconspicuous state variation in history for the emergency event, named the Extro-Spective Prediction (ESP) problem. Based on the proposed dataset, a flexible feature encoder for ESP is introduced to various prediction methods as a seamless plug-in, and its consistent performance improvement underscores its efficacy. Furthermore, a new metric named clamped temporal error (CTE) is proposed to give a more comprehensive evaluation of prediction performance, especially in time-sensitive emergency events of subseconds. Interestingly, as our ESP features can be described in human-readable language naturally, the application of integrating into ChatGPT also shows huge potential. The ESP-dataset and all benchmarks are released at this https URL.
Abstract (translated)
涌现式场景安全是实现完全自动驾驶的关键里程碑，而可靠的时间预测对于在紧急情况下维持安全至关重要。然而，这些紧急情况是具有长期尾性和难以收集的，这限制了系统从获得可靠预测。在本文中，我们构建了一个名为Extra-Spective Prediction（ESP）的新数据集，旨在通过历史紧急情况中的不可见状态变化提供长期预测，名为ESP问题。基于所提出的数据集，为ESP问题引入了一个灵活的特征编码器，作为各种预测方法的无缝插件，并证明了其一致的性能提升。此外，还提出了一个名为clamped temporal error（CTE）的新指标，用于更全面地评估预测性能，特别是在几毫秒的紧迫事件中。有趣的是，由于我们的ESP特征可以用自然语言描述，将ESP集成到ChatGPT中也有着巨大的应用潜力。ESP数据集和所有基准数据集都已发布在https://www.esprite.readthedocs.io/en/latest/index.html。
URL
https://arxiv.org/abs/2405.04100
