A distributed and parallell robot environment for competitive search and rescue using a display wall for visualization

Abstract

In this thesis we present the development of a distributed and parallel environment which offers functionality to robots to support them in their task performance. We want the environment to be a framework where students can experiment with robots, and in which we can arrange robot competitions.

Our motivation for this thesis is an earlier developed robot system at our department. It was used as an instrument to demonstrate the principles and practice of distributed and high performance parallel computing. The system was used by students on advanced courses on cluster architecture and programming, and popular competitions were held in it. The old system had many infrastructure demands and had to be closed. We want to make a new system that has less infrastructure demands and more functionality.

The environment has control over a certain amount of work space where the robots can operate. Within this work space, the environment offers functionality that includes context awareness, location, mapping, naming, and structured interfaces for interaction between the different components. Users can control the robots through the environment, and they assign tasks to the robots. Users can also download extensions to robots through the environment, and robots can upload data to the environment.

The state of the environment, the robots and the work space is visualized on a display wall. Users can interact with this visualization and assign simple tasks through it.