Validation of Robot Model with Mobile Augmented Reality

Abstract

Augmented Reality has excellent potential for many robotic applications, including the validation of robot models in working environments. However, a dedicated depth sensor is only available on a high-end mobile device. Hence, the realistic AR experience for robotic applications is not available for a wide range of mobile phones. This paper presents a way to use a robot control algorithm on a virtual robot model in AR without the need for a depth sensor while keeping the depth required feature such as an occlusion. The connection between the AR and the robot operating system allows the exiting control algorithm to be applied to the augmented robot model. The navigation of a mobile robot with the AR interface was used as an example of robot validation in real-world spaces. Despite some communication delays, the virtual robot can be controlled and navigated in the real world with a certain degree of accuracy. It has been shown that visualizing and controlling a robot in an AR scene can be done with this method. Future work on the interaction between the virtual robot and the real environment should be conducted to expand the application of robot validation with AR.