INTELLIGENT TRAFFIC CONE ROBOT

Authors: LIN HSING-YU, CHANG YU-CHENG, WU CHENG-HAO, CHANG TING-JUI, HO DING-QIAN

School: SRH Hochschule Heidelberg, Dayuan International Senior High School, Taoyuan Municipal Guanyin Senior High School, St. Francis Xavier Junior High School, New Taipei Municipal Taishan Senior High School

Country: TAIWAN

e-mail: eddie@wiipa.org.tw

The Intelligent Traffic Cone Robot is an AI-based transportation safety system integrating artificial intelligence (AI), the Internet of Things (IoT), and autonomous robotics. It is designed for highway accident warning, emergency response, and disaster prevention. Unlike traditional warning triangles that require manual placement, the robot automatically deploys from a vehicle to a safe location, reducing human exposure to hazardous traffic environments. Built on the HUB8735U AI platform, the system uses the YOLOv4-tiny model to detect vehicles, traffic congestion, and road anomalies in real time. RTSP live streaming and OSD functions enable remote monitoring and instant traffic visualization. Once deployed, the robot activates LED warning lights and a buzzer to alert approaching vehicles. For autonomous navigation, vehicle distance is estimated from AI detection bounding boxes, allowing basic following and collision avoidance without dedicated distance sensors. Multiple robots can communicate through Wi-Fi or MQTT to form temporary safety zones, traffic guidance, or construction barriers. By integrating AI perception, autonomous deployment, and multi-robot collaboration, the proposed system transforms conventional warning devices into an intelligent traffic safety platform with promising applications in smart transportation and future smart city infrastructure.