Development, Control Adjustment, and Gesture Recognition of a Quadrotor Helicopter

Abstract

Quadrotor helicopters (quadcopters) have become popular in recent years; because they are simple to operate and steady, they are used in aerial photography, competitive flying, and search-and-rescue missions. In addition, wireless sensors have enabled gesture recognition, and therefore, this study investigated the use of gesture recognition to control a quadcopter. A quadcopter was built using Arduino NANO and MPU-6050 modules, which provided insights into the flight principles of a quadrotor system; furthermore, a proportional integrative derivative controller was modified for steady flight. A Leap Motion device was used to understand the logic behind hand gestures and test the success rate of each gesture, and was then combined with a Parrot AR.Drone 2.0 to achieve gesture recognition controls and uncover problems with connections between hand gestures. These problems were addressed by changing the code in the gesture control program. The results of the completed gesture control experiment demonstrated high gesture recognition performance as well as the ability to meet the requirements for controlling a quadcopter.