Research On Non-contact Manipulator Control System Based On Kinect

With the rapid development of intelligent technology,robot applications are expanding and deepening.The interactive control system based on virtual technology enables the robot to conduct remote control instead of the operator in harsh environments.The friendly human-machine interface is used for human senses and control in special environments.It can create the first perspective of the robot working environment with no one on the scene.Visual sensing technology and wearable body sense technology have developed rapidly in recent years and have been widely used in medical treatment,education and training,animation entertainment,games,exhibitions and other industries.Vision sensors can capture depth images of the human body,while wearable sensors can capture the potential of human muscles as input signals for somatosensory control.In this paper,a non-contact manipulator control system based on "body +" interaction mode is developed by combining the user's natural motion with the robot motion through the fusion of visual sensor and wearable body sense device.The computer takes the spatial position data of the human arm node obtained through analysis and calculation and the wrist movement information obtained through analysis as the control input signal of the mechanical arm,and sends the result to the main controller through wireless communication technology,realizing the real-time mapping of the control robot arm and achieving the purpose of simulating the human arm movement.This paper focuses on the design and construction of a robot arm that imitates the movements and grasping of a single arm.The whole system involves the knowledge of control,software and communication,including the development of visual body sense subsystem and wearable body sense subsystem.Non-contact mechanical arm control system used to access technology and wearable technology,through a computer(PC)after processing,the arm movement sampling signal sent to the robot arm,using C + + Fast Light Toolkit(FLTK)library and Python TkInter library developed two graphical user interface(GUI).C + + FLTK GUI integrates device sampling algorithm,and shows the skeleton tracking sampling data.The Python TkInter GUI gives users the option to run different functions throughout the system.The Kinect sensor is used to capture the user's bone-node data,and the Myo motion control armband is used to interpret gestures,which are further processed to define pulse width modulation(PWM)signals that are used to drive the servo motor on the robot arm.Six servo motors are integrated into the robot arm,three of which are controlled using Kinect data,and the other three are controlled using Myo armband data.A camera module was added to the system and placed behind the robotic arm to provide visual feedback on the computer screen.In addition,all the data and instructions are transmitted wirelessly through wi-fi to achieve remote control of the robot.