Mackenzie Presbyterian University (BRAZIL)
About this paper:
Appears in: EDULEARN21 Proceedings
Publication year: 2021
Pages: 11663-11671
ISBN: 978-84-09-31267-2
ISSN: 2340-1117
doi: 10.21125/edulearn.2021.2441
Conference name: 13th International Conference on Education and New Learning Technologies
Dates: 5-6 July, 2021
Location: Online Conference
Telepresence robot systems incorporate video conferencing equipment into mobile robotic devices that can be controlled remotely. These devices were developed for several areas of use such as: hospital, industrial, commercial, and school. The main purpose of this system is to provide social interaction among humans, specially to children and young that can't go to school because of health issues. The system consists of the physical robot (sensors, actuators, and camera), and an App installed on the user's cell phone allows them to manipulate the robot in their environment, and opens doors to the navigability of many other places and possibilities.

The specific objectives of this work is to address aspects for robotic telepresence in the social sphere, which include:
(1) the mechanical design,
(2) the design of the user interface,
(3) the interaction between the remote user and the user environment
(4) the perception of robotic telepresence systems
(5) on-board Artificial Intelligence improving its usability, navigability, and system security.

The mechanical design was elaborated for facilitating navigability in various environments such as: ascending and descending ramps, passage through narrow places, environments with floor deformity, trajectory change, up to 360 degrees turns within its own axis in a 40cm space. With two wheels the robot has acceleration sensors that control its speed and trajectory, a gyroscope sensor analyzes its position on the X, Y and Z axes, in real time, keeping it always stable even if there is an involuntary bump, providing a safe ride, always keeping the robot on a 90 degree axis, even with a load close to zero the robot will remain stable since the system detects the state of its battery if it is close to zero level and being two safety rods down and keeps it upright. An embedded microprocessor analyzes all sensors and actuators and maintains a strict control over them, while it connects via Bluetooth communication with the tablet, making the interaction between them, and through the commands sent remotely by the manipulator user received on the tablet, the robot moves around its environment. Through an intuitive App installed on the user's mobile phone and another on the Robot's Tablet, the user can move it by means of arrows indicating its orientation and, when touching them, manipulate the robot from a distance, being able to take a picture of the environment and even record them according to future study need. Through cameras and a field vision sensor, the robot is able, using Artificial Intelligence, to recognize its environment, and also learns to navigate, identifying obstacles, ramps, doors, corridors, elevators, and stairs making its navigation better and better. From the robot's learning, it can go from point A to point B without user manipulation and, if it needs intervention, the remotely connected user can change the trajectory. Most telepresence robots available on the market have a high cost, making their usage difficult in schools, and other places. This project presents a viable solution for a low-cost telepresence robot with embedded artificial intelligence, so that people with disabilities and or some comorbidity can attend a classroom, interact with colleagues, and do group work. Finally, the research describes the proposal framework, as well as presents the prototype developed and research challenges of robotic telepresence systems implementations for better social interaction.
Telepresence robot systems, school, health issues, artificial intelligence.