Remote Learning and Operation of Robotic Cells

פרויקט במסגרת קול קורא 2 לקידום יישומי טכנולוגיות למידה

  

Ofir Goldstain, Irad Ben-Gal, Yossi Bukchin

Department of industrial engineering, Tel-Aviv University

 

In this work [1] we propose a method to integrate remote control and manipulation of robotic work cells (Tele-Operation) with aspects of remote learning. The motivation for this integration is to enable students to practice not only the remote activation of a robotic cell, but also the availability of redesigning and optimizing both the work cell environment and the work plan itself.

Remote control and manipulation of robots has been previously used to perform predetermined tasks, often in a hostile, unsafe, inaccessible or remote environment. These tasks usually require the user's focus on the remote manipulation of the robots rather than on the optimization of the work plan. Accordingly, the interfaces for these tasks were designed to deal with a given environment setting and a set of fixed tasks.

Remote learning, on the other hand, focuses on the ability to teach the user how to obtain an optimal solution given a flexible set of tasks and tools; it has been an increasingly growing field in the last two decades. Internet development, as well as the increase in PC's capabilities and bandwidth capacity, has made remote learning through the internet a convenient learning preference that lead to a variety of new learning interfaces and methods. In relation to the tasks considered here, with the development of technology it is now possible to combine 3D simulation tools with a Remote Control and Manipulation interface, to comply with more complicated tasks in flexible robotic cells.

The specific problem we address in this work is how to enable a number of students to simultaneously design and optimize a work plan for a remote robotic cell. The students are required to learn, through a "try and error" process, the appropriate way to operate the robotic cell previous to its remote online activation. The learning schemes that we found most efficient, in terms of learning time and efforts as well as in the resource usage, is having the students operate a simulation module before the online execution. The proposed system approaches remote robotic manipulation from a new angle, enabling students to learn and deal with different robotic cells as well as experiencing the control and manipulation of such cells from a remote site. The learning process is based on simulation and optimization of tasks that require a cognitive learning process, which differs from the one required for a motoric learning when operating a robotic cell on site.

We have developed an internet web-site interface and integrated it with a particular 3D simulation program to be installed on the user's computer. The developed website enables us to administrate the students and control the system usage, while keeping the work site's computer invisible to the end user, thus, preventing unnecessary safety problems. The integrated simulation tool supports redesign of the layout as well as a "tryal and error" process, which are vital for learning and skills improvement. Bandwidth required for this type of solution was found to be significantly lower from the other considered approaches. The web-site we present can later be used to further integrate full scale remote learning laboratories in collaboration with other organizations.

 

[1] The work was presented at the UMTIK 2006 conference, Turkey. It was accepted for publication in the European Journal of Engineering Education.