Design and Development of a Rescue Robot to Identify Human Presence in Disaster Scenarios by Using Artificial Intelligence Assisted Sensor Support

Abstract

In disaster scenarios, rapid and accurate identification of human presence is critical for effective rescue operations. This study presents the design and development of an AI-assisted rescue robot equipped with advanced sensor support for detecting human presence in various disaster environments. The proposed system integrates multiple sensors, including thermal imaging, LiDAR, ultrasonic sensors, and cameras, to gather comprehensive environmental data. A YOLO-based AI model processes this data to identify human presence with high accuracy. The model was trained and validated on diverse datasets simulating real-world disaster conditions. Results show that the proposed model outperforms nine existing models, achieving an average accuracy of 97.3% across four different disaster scenarios. This accuracy is significantly higher compared to the closest competing models, highlighting the robustness and reliability of the proposed system. Additionally, the model demonstrated a precision of 96.7%, a recall of 96.5%, and a response time of 49.5 ms, all of which contribute to its superior performance in time-sensitive and complex environments. The findings suggest that this AI-assisted rescue robot is a promising solution for enhancing the efficiency and effectiveness of search and rescue operations in disaster scenarios.