Wheeled robot control based on gesture recognition using the Kinect sensor

Abstract

Human Machine Interaction (HMI) plays a very important role in intelligent service robot research. Traditional HMI methods such as keyboard and mouse cannot satisfy the high demands in some environments. To solve this problem, many researchers pay attention to vision-based gesture recognition research recently. This paper proposes a simple method to control the movement of a robot based on Kinect which provides skeleton data with low computation, acceptable performance and financial cost. The method can recognize eleven defined gestures by using the coordinates of joints, which are obtained from the skeleton model provided by the Kinect SDK. A Khepera III robot is used as a prototype control object, to verify the effectiveness of the proposed method. The experimental results show that the success rate of gesture recognition is over 96%. The proposed method is robust to work in real-time.