Development of a Gesture Detection System for the Control of a Robotic Arm

American Journal of Science and Technology

OsahorUche¹  L.O. Kehinde¹ 

[1] Department of Electronic and Electrical Engineering, Obafemi Awolowo University, Ile-Ife, Nigeria.

关键词: Gesture Detection;    Kinect Sensor;    Graphical Interface and Robotic Arm;   

学科分类：工程和技术（综合）

来源: AASCIT

【 摘 要 】

This study developed a gesture detection system for the control of a robotic arm. This was achieved with the view of developing a simple and user friendly approach for robotic arm manipulation. A Microsoft Xbox™ Kinect sensor was used as the input device for retrieving real-time joint information from the user. Kinematic algorithms were developed by applying Denavit-Hartenberg parameters and geometric equations. The acquired gestures from the human subject were used to carry out various kinematic routines as deemed fit by the user. A graphical interface was also developed to provide real time feedback to the user. The robot was able to execute basic gestures as instructed by the user rated at about 80% success rate. Most gestures developed were effective enough for the scope of this project. However, minimal gesture detection errors were recorded averaging below 20%. A standard deviation error of 3.8 at a variance error of approximately 14.1 was recorded. The study implemented an effective gesture system that was capable of executing the basic routines required for robotic arm manipulation in real time.

【 授权许可】

Unknown