Design and Implementation of Platform of Educational Mobile

Design and Implementation of Platform of Educational Mobile Robots Based on ROS Xinguo Yu, Lvzhou Gao, Bin He, Xiaowei Shao May 20, 2017 National Engineering Research Center for E-Learning Central China Normal University

Outline • Research subject • Problem • Related works • Experiments and evaluation • Conclusions

Research subject • Now we have a research subject to build a education robot who can provide services for teaching activities in classroom, and the major functions of robot include four parts: 1, Speeking. The robot could talk with students through voice. 2, Autonomous Movement. It can go anywhere in classroom without help from human. 3, Object following. Using computer vision recognition one’s body, robot follows him/her at a certain distance. 4, Indoor localization. Project Multimedia sources on the wall accurately based on the accurate localization.

Problem We need a robot to test our algorithm and to adapt to our classroom. • In recent years, several mobile robotic platforms have emerged, for instance NAO, Turtlebot. However, there are still problems, such as high price, weak capability of redevelopment, and poor extensibility, which have seriously hindered the application and promotion of educational robot. • Lower price is a significant factor for wildly accepting by school and students. At present, robot platform is expensive generally, in addition, this platform can’t meet many special need of education. • We want to have our own robot system including hardware and software, which contribute the formation and promotion of education robot.

Related works • Selecting robot operating system Projects Windows Android Ubuntu ROS Openness No open source Open source Software independence Poor Common High Difficulty of development Common Low High Low Real-time Fast Common Slow Fast(ROS 2. 0) Stability High Common Application areas Medical field commerce research, education, commerce

Related works • Hardware Framework Design of Educational Robot

Related works • Design of Robot Software Framework

Experimental results u Control of Motion error Error statistics of linear velocity of robot 0. 3 m/s Displacement/ cm Measurement times Average speed/ (m/s) Feedback speed/ (m/s) Error/ (m/s) 100 300 500 700 900 20 20 20 0. 301 0. 300 0. 301 0. 302 0. 303 0. 001 0. 002 0. 003 0. 002 Error statistics of robot 6 rad/s angle Sequence Target rotation angle /(°) Actual rotation angle /(°) Feedback angle /(°) Error/ (°) 1 90 90. 8 90. 2 0. 6 2 180. 7 180 0. 7 3 270. 9 270. 2 0. 7 4 360 361. 3 360. 4 0. 9

Experimental results u Sl. AM and Navigation Autonomous navigation of robots based on map is an important solution for robot to move autonomously in teaching scene. The prototype machine integrated three commonly used function packages for robotic autonomous navigation: move_base, gmapping, amcl, which are provided by ROS.

Conclusions and future work • Taking the application of robots in education and the needs of educational robot research into consideration, we designed a three-layer architecture of educational robot, as well as a hardware framework and a software framework. Then, we implemented an embedded educational robot prototype by transplanting ROS into a Raspberry Pi with good performance and lower price. In the end, experiments of motion control, speech and vision of robot were carried out on the prototype. • we will optimize the existing educational robot prototype. On the other hand, we will focus on the development of the subject-oriented robot solution, so that the robot will become a mobile "teacher".

Talent recruitment • Recruit talents in ICT to do Rn. T. High requirement and high payment for Joint Institute btw CCNU & UOW. • Maser, doctor, enginner, teacher who are skilled in communication, mechanical automation, and robot system integration, etc. • CCNU web: http: //www. ccnu. edu. cn/ • NERCEL web: http: //nercel. ccnu. edu. cn/ • Professor Yu——http: //nercel. ccnu. edu. cn/info/1072/3066. htm • Professor yu Email: 2429346468@qq. com

Email: gaolvzhou@outlook. com