Landmine Detection Robot Car Using Machine Learning

Abstract

Landmine detection remains a critical challenge in ensuring security in post-conflict regions. This study presents an innovative autonomous landmine detection system that integrates a robotic vehicle with IoT devices and machine learning-driven real-time image analysis. The proposed system features a NodeMCU-based robotic platform powered by an L298N motor driver, which can be remotely operated using the Blynk app. An ESP32-CAM module is outfitted on a robot vehicle that captures images of the terrain; manipulated into binary data and analyzed using a pre-trained model of machine learning. This domain model trained on a dataset of terrain images will identify landmines by recognizing specific visual patterns.

The robot independently moves and processes real-time images, hence providing a scalable and efficient solution to landmine detection in risky areas. FW for the ESP32-CAM module will be uploaded and controlled with FTDI technology in seamless communication and control. The use of such a system for autonomous navigation and detection enhances the safety and execution of demining activities, and as such, will significantly reduce human exposure to danger. This research shows what a powerful role IoT-driven robotics and artificial intelligence can play in addressing urgent humanitarian challenges.