Regresar

Hand Gesture and Arm Movement Recognition for Multimodal Control of a 3-DOF Helicopter

Abstract:

This paper presents the application of hand gestures and arm movements to control a dual rotor testbench. A multimodal control method is developed for a 3-degrees-of-freedom (DOF) tandem helicopter based on surface electromyography sensors and an inertial measurement unit (IMU) included in the Myo Armband sensor. The recognition system can classify five different hand gestures which are used for switching between flight modes and generating set point values for the helicopter. The 3-DOF helicopter testbench is fully designed and implemented as a low cost alternative for assessing the effectiveness of flight controls for unmanned aerial vehicles. The position of the helicopter is regulated by a cascade-dual-PID control scheme that allows a fast switching between a gesture mode and an IMU mode. Experimental results show the effectiveness of using hand gesture recognition and arm movement for controlling an aerial test bench with a fast and accurate response.