Vision-based Autonomous Landing System for Quadcopter Drone Using OpenMV

Nowadays, drones have a wider variety of applications. The drones are now used for daily activities that the demand for drones is increase and the shift to a lower price. Then, drones manufactured with poor build quality. Consequently, the drone is now more vulnerable to crashes. The crash on landing is one of the most common errors made by novice users. This research utilizes a vision-based precision landing method for quadcopter drones equipped with low-cost cameras capable of tracking AprilTag. The system deploy the Pixhawk as the flight controller, Mission Planner as the Ground Control Station (GCS) software, the OpenMV H7 camera, telemetry, and a computer. The test results obtained are the outdoor accuracy for 2 m altitude is outdoors 8.94 cm, semi-outdoor 9.1 cm, and indoor 11.7 cm. The average accuracy for 3 m altitude outdoors is 9.35 cm, semi-outdoor 9.45 cm, and indoor 10.8 cm, which indicates the system can perform vision-based autonomous landing for quadcopter.