Improvement of Collision Avoidance in Cut-In Maneuvers Using Time-to-Collision Metrics

TL;DR

This paper introduces a novel collision avoidance strategy for autonomous vehicles in cut-in scenarios, combining deep learning with Time-to-Collision metrics to improve prediction accuracy and evasive action planning.

Contribution

The paper presents an innovative integration of deep learning with TTC metrics specifically designed for challenging cut-in maneuver scenarios in autonomous driving.

Findings

Enhanced collision prediction accuracy in cut-in scenarios

Improved evasive maneuver effectiveness over traditional TTC methods

Demonstrated robustness of the approach in simulated environments

Abstract

This paper proposes a new strategy for collision avoidance system leveraging Time-to-Collision (TTC) metrics for handling cut-in scenarios, which are particularly challenging for autonomous vehicles (AVs). By integrating a deep learning with TTC calculations, the system predicts potential collisions and determines appropriate evasive actions compared to traditional TTC -based approaches.