Reinforcement Learning for Predicting Traffic Accidents

TL;DR

This paper introduces a reinforcement learning approach using the DARC method for early traffic accident prediction from dashcam videos, achieving earlier and more precise forecasts to enhance autonomous driving safety.

Contribution

It is the first to apply the DARC reinforcement learning model to the task of early traffic accident prediction from dashcam footage.

Findings

Predictions are made 5% earlier on average.

Improved multiple metrics of precision over existing methods.

Demonstrates the potential of RL to enhance autonomous driving safety.

Abstract

As the demand for autonomous driving increases, it is paramount to ensure safety. Early accident prediction using deep learning methods for driving safety has recently gained much attention. In this task, early accident prediction and a point prediction of where the drivers should look are determined, with the dashcam video as input. We propose to exploit the double actors and regularized critics (DARC) method, for the first time, on this accident forecasting platform. We derive inspiration from DARC since it is currently a state-of-the-art reinforcement learning (RL) model on continuous action space suitable for accident anticipation. Results show that by utilizing DARC, we can make predictions 5\% earlier on average while improving in multiple metrics of precision compared to existing methods. The results imply that using our RL-based problem formulation could significantly increase the safety of autonomous driving.