Real-time hand guiding of industrial manipulator in 5 DOF using Microsoft Kinect and accelerometer

Human computer interaction is currently developing toward more intuitive and natural ways of communication. The main goal of research presented in this paper is the development of a system for controlling a robot manipulator in 5 DOF using human motions. The developed system reads and interprets sensor data from a Microsoft Kinect and an accelerometer embedded in a smartphone and use this data and the robot kinematics to generate a position and orientation reference to the robot controller allowing real-time interaction with the robot. The system has been implemented and tested on a Universal Robot manipulator arm with 6 joints. This robot has an embedded controller able to receive a reference in both Cartesian and joint space and calculate and track trajectories that fulfills this reference. The main result of this research is the architecture of an industrially oriented hand guiding system.