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

TL;DR

This paper introduces the Extro-Spective Prediction (ESP) dataset and a novel feature encoder for long-term emergency scenario prediction in autonomous driving, along with a new evaluation metric, enhancing safety predictions.

Contribution

The paper presents a new dataset, a flexible feature encoder, and a comprehensive evaluation metric for long-term emergency prediction in autonomous driving.

Findings

Feature encoder improves prediction performance across methods.

The CTE metric offers better evaluation in time-sensitive scenarios.

ESP features are human-readable and integrate well with ChatGPT.

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 https://dingrui-wang.github.io/ESP-Dataset/.