Towards a CloudSim-Based Learning Environment

CloudSim [1] is an open source web application used to deploy cloud based machines, currently through Amazon Web Services [2], that provide open source robotics tools such as ROS [3] and the Gazebo Simulator [4]. Being web based, CloudSim allows users to interact with robot simulations easily, reducing issues associated with both computer hardware and software compatibility. In effect, CloudSim allows its users to learn or try out open source robotic software with little need to setup their own system. This is especially useful for the curious hobbyist or in schools or universities that do not have a dedicated robotics laboratory to install ROS and Gazebo. An instructor or administrator with an AWS account can then use CloudSim to launch Gazebo simulations while learners can use gzweb [5] to view the simulation and IPython notebook  [6] to interact with the simulated models from a web browser. Alternatively, SSH access to the machines are also available.

While CloudSim can manage launching the machines for its users, data from the simulation session is lost after the CloudSim machines are terminated after use. Data, in the from of task completion percentage or final score would provide learners a way to keep track of their progress in a course, simulation task or tutorial.

Several modules are needed to create a learning environment based on CloudSim; the learning portal that contains the a list of tutorials and learning challenges and tracks user progress and the simulation files associated with the tutorials and challenges. Google CourseBuilder 1.5.1 [7] will be used to develop a prototype for the portal and with ROS (Hydro for Development and Groovy for Deployment) and Gazebo 1.9 to develop the simulation tasks and score system. The posts below will document the project development:

  1. Populating Gazebo Worlds
  2. On Gazebo plugins
  3. Customizing GCB for CloudSim
  4. Deploying Gazebo Simulations in CloudSim
  5. Final Thoughts

The project repository can be found here:


Tags: , , , ,