Social Robots for Road Safety: Pedestrian Crossing Assistance Use-Case

Abstract

Today, robots are being commonly used in industry, healthcare, household, and many other fields, but not in intelligent transportation systems. We aim to fill this gap by adapting the humanoid-like social robot ARI by enhancing its perception, decision-making, and interaction capabilities to manage complex traffic scenarios effectively. Key advancements include upgraded sensors, YOLO-based perception algorithms, and improved mechanisms to interact with pedestrians and vehicles. The robot establishes its ability to detect pedestrian intent, evaluate vehicle behavior, and facilitate safe crossings. This proof-of-concept presents a step toward the integration of robotic systems into traffic management, contributing to improved safety and efficiency in challenging traffic environments.