The Design And Implementation Of Control System For Robotic Arm Human-computer Interaction Based On Position And Gesture Recognition

With the development of robot technology, robot has become an important part of high-tech application fields. This article topic source project "The key technology research on Human-Robot Interaction of space wearable computing" aimed at the research of space robot wearable Human-Robot Interaction way, robotic arm is an important part of all kinds of robot, the robotic arm control effect directly affects the availability and efficiency of the robot. Therefore, the source project takes robotic arm as object, studied the Human-Robot Interaction.In this thesis, a robotic arm control system based on Position and Gesture Recognition is designed and implemented. In some areas, due to the particularity of operator and robot working environment, Requirements control system to facilitate operators operation and less dependent on the environment. And because of the existence of long-distance transmission delay, we should try to reduce the control frequency. Described in this paper, the system adopts the wearable attitude sensors calculating control information; less affected by the environment and has the wearable which can liberate the operator’s hands. The system design and implements a control mode with coarse and fine tune. Firstly, use coarse adjustment to make the robotic arm near the destination location within fewer control frequency. Secondly, fine-tune it again. This control method reduces the control frequency of the robotic arm effectively. The main work is completed as follows:1. The Position and Gesture Recognition system is implemented. The system gets more stable data by processing the angular velocity and acceleration information, and fuses the data to calculate the gesture angle; calculate the space position coordinate of the wrist according to the gesture angles of the shoulder, elbow and wrist as the coarse adjustment information; Set eight basic action with small angle according to the changes of the wrist’s gesture angle as the fine-tuning information.2. Map the wrist to the end of the robotic arm and set the wrist’s coordinates as the robotic arm’s purpose coordinates; Improve the traditional method of calculating inverse kinematics to get the rotation Angle of each joint; expounds the trajectory planning method.3. The demo interface based on MFC is designed and implemented, Which has the following functions: A virtual robotic arm model is established in Open GL environment and displayed in the view window; A module of kinematics simulation is embedded to drive the virtual robotic arm model complete graphic simulation; Generate manipulator control instruction according to the rotation angle of the virtual robotic arm model; The control instruction, the virtual robotic arm model’s joint angle and the Position and Gesture of the end of the virtual robotic arm model are displayed in the display window; Provides the speech interface to complete the communication with the speech control part.4. An experimental platform is developed which demonstrate the naturalness, feasibility and the reliability of the system.Through a series of functional tests and verification, we find the system meet the design requirements basically. And the experimental results show that the system effectively reduces the control frequency and improve the control efficiency.