Deep Reinforcement Learning for Advanced Longitudinal Control and Collision Avoidance in High-Risk Driving Scenarios

TL;DR

This paper presents a deep reinforcement learning algorithm for vehicle longitudinal control and collision avoidance that considers both leading and following vehicles, effectively preventing pile-up collisions in high-risk, dense traffic scenarios.

Contribution

It introduces a novel deep reinforcement learning approach that accounts for multiple vehicle behaviors, improving safety in complex traffic situations.

Findings

Successfully prevents pile-up collisions in simulations

Effective in emergency braking dense traffic scenarios

Handles heavy-duty vehicle interactions

Abstract

Existing Advanced Driver Assistance Systems primarily focus on the vehicle directly ahead, often overlooking potential risks from following vehicles. This oversight can lead to ineffective handling of high risk situations, such as high speed, closely spaced, multi vehicle scenarios where emergency braking by one vehicle might trigger a pile up collision. To overcome these limitations, this study introduces a novel deep reinforcement learning based algorithm for longitudinal control and collision avoidance. This proposed algorithm effectively considers the behavior of both leading and following vehicles. Its implementation in simulated high risk scenarios, which involve emergency braking in dense traffic where traditional systems typically fail, has demonstrated the algorithm ability to prevent potential pile up collisions, including those involving heavy duty vehicles.