Deployment and Optimization of Wireless ad-hoc Communication Networks

Wireless communication is a key technology in Distributed Intelligent Systems. Robotics applications pose challenges to current technology due to increased mobility and a large variety of throughput and latency requirements.

We have developed a platform for autonomous deployment of a mobile communication backbone, which can provide networking capability to other robotic agents, sensor networks, or people. We are experimenting with a team of 10 iRobot Create robots, Atheros embedded Linux 802.11g radio, accelerometer and GPS. The system can be programmed using the swarm programming language MIT Proto.

The algorithms and hardware design are particularly motivated by a low-cost solution, which achieves robust behavior using only scarce (or none) positional information. Building up on a fully reactive algorithm that requires no positional information, we incrementally develop algorithms with better performance that rely on more and more accurate sensor information.

We developed a fully distributed, reactive algorithm for deployment and maintenance of a mesh network. The algorithm has minimalist requirements on the individual robotic node, which are limited to wireless signal strength estimation and bumper sensors. This makes the proposed solution suitable for deployment of large numbers of comparably cheap mobile communication nodes. Robots explore the configuration space by random walk and stop only if their current location satisfies user-specified constraints on connectivity and network topology. Resulting deployments are robust and convergence is analyzed using both kinematic simulation with a simplified collision and communication model as well as a probabilistic macroscopic model.