INTELLIGENT HUMAN RESCUE ROBOT: AN AI-DRIVEN APPROACH WITH FPGA TECHNOLOGY

This study introduces an innovative AI-driven robotic system designed to enhance search and rescue operations in disaster-stricken areas, including earthquakes, floods, and conflict zones. The system integrates a multi-modal sensory framework combining RGB and thermal imaging, processed through a robust pipeline: raw sensor data undergoes validity checks and noise filtering to ensure input integrity, followed by human detection via the You Only Look Once (YOLOv8) algorithm. Concurrently, ultrasonic sensors relay environmental data to a Field-Programmable Gate Array (FPGA) module, enabling real-time obstacle avoidance and dynamic path planning for rapid, precise navigation. Experimental results demonstrate a 96% accuracy rate in survivor detection, outperforming comparable AI-FPGA hybrid systems—particularly in identifying partially obscured humans. The system also exhibits superior response times and operational safety, as quantified in the results section. Future work will focus on expanding sensory modalities, adaptive learning mechanisms, and multi-robot coordination to further enhance mission efficacy