Autonomous Agricultural Robot with Inductive sensors for Real-Time Human Detection

Abstract

This paper presents the development and verification of a real-time human detection system designed for autonomous agricultural robots. The system aims to enhance safety by preventing collisions between robots and human workers in the field. Traditional sensor-based detection methods like vision, infrared, and ultrasonic sensors have limitations in outdoor agricultural environments due to lighting, weather, and obstructions. To address these challenges, the authors propose a system utilizing permanent magnets and inductive sensors. The robot is equipped with a strong magnet, while human workers carry sensors that detect magnetic fields. When a dangerous proximity is sensed (magnetic field exceeds 1 mT), the system wirelessly sends an emergency stop signal to the robot. A prototype was developed using a radio-controlled excavator, microcontroller, and magnetic field sensors, and experiments confirmed successful detection and stopping at a safe distance of 60 cm. Finite element method (FEM) analysis further determined that a 25 kg magnet is needed to generate sufficient magnetic flux density for detection. This system provides a reliable safety solution for human-robot collaboration in agriculture, even under adverse environmental conditions.