The open-source swarmbot concept continues to spread.
We've touched on the subject before, but a recap may be fruitful: in order to better understand changing local environmental conditions (for agriculture, conservation, health, etc.), it helps to have a multiplicity of sensors providing data streams; those sensors can cover more area if they're mobile; rather than having to control each individual mobile sensor, "swarming" or "flocking" behaviors allow the bots to position themselves to maximize coverage yet retain local communication; by making the project free/open-source, people in low-income or resource-restricted communities can still take advantage of the system's capabilities.
The Flockbot Project, at the computer science department of George Mason University, is an attempt to design mobile, swarming robots able to perform useful actions, all at a (relatively) low cost.
This website describes an open design for a small, $800 robot suitable for "swarm"-style multiagent research, robotics education, and other tasks. Our goal is to get as much functionality as possible from $800 per robot, replicate the robot many times to create a small collaborative swarm, and document the results to make it easier for you to do the same. We hope to foster collaboration in the wider community and, ultimately, lower the entry-level costs for building such robots.
The robot design is remarkably complex, given the limited resources. It combines a Linux-based computer, wireless networking, a camera, a gripper, and multiple IR sensors, all on a 7" diameter wheeled platform. Future modifications include a move to a smaller control system, better mobility, and a price cut to below $500.
The Flockbots site focuses on the design and construction, with little information about actual experimental use. But the online materials are more than sufficient for hobbyists and hackers to follow in their footsteps. Who will be the first to use garage robot swarms in the field?