Multi-Vehicle Unmanned Following Operation Driving System Based on OpenMV

Abstract

In response to the low level of agricultural production automation and labor shortage, to improve the efficiency of agricultural machinery utilization and the reliability of automatic navigation, this paper designs an intelligent agricultural machinery system based on OpenMV for multi-vehicle unmanned following operations. Firstly, the kinematics of the agricultural machinery vehicle is analyzed, the path recognition and location line extraction are carried out using the OpenMV vision module, and the agricultural machinery steering control system is designed using the cascade PID closed-loop control technology, which achieves accurate steering control and path tracking on straight lines and curves. Subsequently, using embedded microcontrollers to manage the workflow, utilizing OpenMV visual modules to receive, store, and process positioning information, and sending control instructions to execution units such as the TB6612FNG motor drive module. Moreover, in response to the communication issues involving multiple HC-05 Bluetooth modules in the system's multi-vehicle communication, corresponding protocol specifications have been developed, and combined with the use of the HC-SR04 ultrasonic module to maintain dual vehicle distance and obstacle detection, achieving multi-vehicle collaborative following operations. The experimental results show that the system achieves a path recognition navigation accuracy of 93.4% and a communication success rate of 97.5% in simulating the process of farm harvesting and transportation, effectively proving the multi-vehicle collaboration of the system in the paper