Three Degrees of Freedom Robot Arm with Augmented Reality Markers for Education and Training Courses in Robotics

Abstract

Robotics with artificial intelligence is an essential basis of industry 4.0, where intelligent robots with full-time connectivity react to flexible manufacturing, sharing workspaces with working employees and other machines. Designing these robotic systems with artificial intelligence and applicable to all manufacturing processes implies knowledge of experts in computer science, physics, mechanics, and industrial processes, among others whose previous knowledge in robotics is minimal. The lack of space perception in a three-dimensional world is a common difficulty for those who start working with robots, and missing attendance to this problem can produce several injuries and high-cost damages. Alternatively, employees and designers can work with augmented reality robots to solve this problem since Augmented Reality is another critical technology in industry 4.0. However, students must work using basic libraries to solve the pose estimation problem and render three-dimensional objects by applying the mathematical background of robotics and Augmented Reality instead of commercial libraries that deal with the situation as a black box solution. This alternative lets the students understand the background problem and design their new technologies. In this work, we propose using an Augmented Reality with markers for rendering a robot arm that shows frames, moving direct, and inverse kinematics once the students provide the Denavit–Hartenberg convention that correctly describes the robot. First, we estimate the robot pose by determining the homography matrix between the initial position and the current frame acquired from the camera. Then we modify the position and orientation of the rendered robot with all its frames and graphical information. Finally, we render the robot in the Webots environment to simulate robots.