ESP-Dataset

Extro-Spective Prediction for Long-term Behavior Reasoning in Emergency Scenarios

Dingrui Wang1,4, Zheyuan Lai1, Yuda Li1, Yi Wu2, Yuexin Ma3, Johannes Betz4, Ruigang Yang1, Wei Li1
1Inceptio Technology, 1Nanjing University of Posts and Telecommunications, 1ShanghaiTech University, 4Technical University of Munich
Paper arXiv Video Code Data

The autonomous truck was forced to brake suddenly as a sedan cut in front of it on the highway

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.

Video

BibTeX

@article{dingrui2024esp,
  author    = {Dingrui Wang, Zheyuan Lai, Yuda Li, Yi Wu, Yuexin Ma, Johannes Betz, Ruigang Yang, Wei Li},
  title     = {ESP-Dataset: Extro-Spective Prediction for Long-term Behavior Reasoning in Emergency Scenarios},
  journal   = {ICRA},
  year      = {2024},
}