Unmanned Robotic Online Laboratory with an Intelligent Cloud Teacher

One-on-one student-instructor communication is essential in many aspects for science and engineering education. Especially under and post COVID-19, individual student-instructor interactivity becomes increasingly precious yet growingly difficult. To address this issue, this poster paper develops a so-called "cloud teacher", which is an intelligent conversational agent for answering the questions from students in building and manipulating their robots online. The basic underlying idea is to train the agent by utilizing open-source artificial intelligence tools from Google's DeepMind, such that the agent can understand and answer the questions raised by students in making their robots online. Specifically, distinct features of cloud teacher include: 1) Provide instant textual answers to students' questions regarding robot assembly and computer programming (Arduino and SolidWorks: Two most common software in robot control and 3D printing). 2) Supply video demonstrations for helping students in building and manipulating their robots. 3) Automatic assessment of students' online lab performance on controlling their robots. 4) Personalized quizzes based on big data analytics of students' historical questions and performance. Based on the above features, an unmanned robotic online lab has been constructed with 11 sets of real robotic arms, controllers, touchable monitors and corresponding software. The approach of Al powered conversational agent has demonstrated its effectiveness in 4 engineering courses with more than 300 students. Furthermore, we collaborated with an oversee university, and applied it in the distant robotics education with 50 students from U.K. This method has a high potential to be explored across different disciplines and countries for virtual experiential learning in the new normal.