SPHERESCUE: Design and Development of Search and Rescue Robot with Integrated First Aid Supply

Abstract

This study aimed to develop SPHERESCUE, a low-cost, remotely controlled spherical search and rescue robot with integrated first aid storage, live video, and GPS capabilities. Through remote operation and AI-enhanced human detection, the project seeks to address the danger rescuers face in hazardous terrain by providing a safer means to locate and assist survivors. The prototype was designed and constructed using Arduino Uno, ESP32-CAM, NEO-6M GPS, and a servo-controlled hatch for first aid deployment. Its spherical body, inspired by DOST’s Project Sphere and Star Wars’ BB-8, was made from papier-mâché reinforced with lightweight internal support. Programming was done in C++ via Arduino IDE to control movement, navigation, and detection systems. The prototype was tested on various terrains to measure stability, navigation efficiency, and accuracy of live video and GPS telemetry. The first aid delivery mechanism was also tested through multiple trials to assess success rates in opening, accessibility, and security of stored supplies. SPHERESCUE’s results showed efficiency in navigating flat and uneven terrains. Although with minimal errors in identifying targets in transmitted live video using AI-assisted detection, it also lacks maintaining stability on inclined areas. The first aid hatch achieved a 60% success rate, indicating functionality with room for improvement. SPHERESCUE demonstrates strong potential as a low-cost, accessible robotic aid for safer and faster disaster response. As a prototype, limitations such as terrain adaptability and hatch precision are expected.