Cooperative Behaviour of Heterogeneous Mobile Robots

The diversity of mobile robots has increased significantly during the past decades. And since many complex applications have different requirements on sensors and robots, clearly not every robot is suited for every task. This is the main reason that makes teams of heterogeneous mobile robots attractive. The employment of different types of robots has the great advantage to exploit the robots’ individual strengths.

A team of heterogeneous mobile robots

At our institute,we work with two kinds of robots in this task: parent robots with state-of-the-art computation power and capable sensors, as well as multiple smaller and more flexible child robots with restricted sensing and computational capabilities. The parent robot is a custom built service robot with a height of approx. 1.5 m. It is equipped with a laser scanner and an omnidirectional vision system that allows for detecting and tracking the red hats of the small mobile robots. The 13 child robots of our lab are in the size of approx. 15 cm and were developed by the German computer magazine c’t (www.ct-bot.de). They possess only the restricted computation power of a microcontroller and low-cost sensors.

 

Central research questions are:

  • Multi-robot coverage: Area coverage is an essential task for mobile robots that finds its relevance in various applications, e.g., floor cleaning, de-mining or harvesting. Generally, in coverage, robots visit every point in a target area, at least once.
  • Cooperative mapping: Mapping an environment is a first step allowing a robot to afterwards self-localize in it. We develop cooperative mapping techniques in teams of heterogeneous mobile robots.

 

Videos concerning this topic:
See our current robotic videos in our Youtube channel.

 

Publications of our institute concerning this topic:

[1] Marius Hofmeister, Marcel Kronfeld, and Andreas Zell. Cooperative visual mapping in a heterogeneous team of mobile robots. In IEEE International Conference on Robotics and Automation (ICRA), pages 1491--1496, Shanghai, China, May 2011. [ pdf ]
[2] Marius Hofmeister and Marcel Kronfeld. Multi-robot coverage considering line-of-sight conditions. In 7th IFAC Symposium on Intelligent Autonomous Vehicles (IAV), Lecce, Italy, September 2010. [ pdf ]

 


Contact

Marius Hofmeister, Tel.: (07071) 29-78988 marius.hofmeister at uni-tuebingen.de