Building a mobile robot testbed for multi-agent and distributed control applications


Staff Mentor:

dr.ir. T. Keviczky (Tamas)


Other Mentor(s):

A. Simonetto

Keywords:

Multi-agent systems; Distributed control

Description:

The problem of coordination and autonomous operation of groups of vehicles represents challenges that are becoming increasingly important to address for enabling unprecedented efficiency and technologies in diverse application areas such as transportation networks, logistics, high-performance agricultural systems, space exploration and mobile sensor networks. These areas require new solutions for distributed mapping and path planning, information sharing and consensus, just to name a few. Researchers are currently testing these algorithms in robot soccer teams, autonomous underwater vehicles, and unmanned aerial and ground vehicles as well.

The aim of this MSc project is to build a mobile robot testbed in an area of approx. 5 x 5 meters, starting from off-the-shelf components in order to test several distributed control algorithms.

First, an appropriate hardware platform has to be chosen and interfaced with a desktop PC for remote control. This task requires both hardware and software skills. Second, a suitable localization techique has to be chosen and implemented, with particular attention to the hardware and software limitations of the experimental setup and the lab area. These choices and developments will have a fundamental effect on the testbed's capabilities. Finally, a user-friendly Matlab/Simulink interface has to be developed, which simplifies the task of running experiments and allows easy access and iteraction for researchers interested in testing various algorithms on the testbed.

We seek an enthusiastic MSc student for this exciting and challenging project, who has some background in hardware/software design and is interested in developing tools to control a group of mobile robots autonomously.


© Copyright Delft Center for Systems and Control, Delft University of Technology, 2017.