Dual-hand gesture controlled quadcopter robot

Abstract

The quadcopter aircraft has been widely used in many fields today. However, the complicated and professional operation of the quadcopter still needs to be qualified by professionals. How to simplify the operation control is one of the urgent problems in the further development of the quadcopter aircraft. Based on the kinect sensor, a new simplified operation scheme is proposed, and a dual-hand controllable quadcopter robot is designed and extended to carry out the task of grasping and transporting object. In the proposed scheme, firstly we obtain the 3D video stream data by kinect. Secondly, we perform noise pre-processing on the depth image to extract the stable skeleton node data. Thirdly, the skeleton node data is used to establish the control signal reference coordinate system, and a dual-hand gesture information is mapped into the control information of the quadcopter and the mechanical claw. Finally the hardware system of PSoC as the core device and the software control system are designed. The experimental results show that the quadcopter robot can achieve stable control of the quadcopter attitude by dual-hand gestures, and use mechanical claw to grasp and convey a object by observation. The control method is simple and effective.