Slashdot Mirror
Linguistics Could Help Future Driverless Cars Cooperate Better (thestack.com)
An anonymous reader writes: A team of swarm robotics researchers have applied a linguistics technique typically used in manufacturing to automatically program and control a 600-strong robot fleet. The scientists found that human error was significantly reduced, making the solution safer and more reliable than previous 'trial and error' approaches. The tasks in the experiments were defined by a graphical tool, which a machine then automatically translated to the bots. The supervisory technique uses a linguistics system through which the robots construct their own 'words', related to what they can 'see' and which moves they choose to action next. Robots will only perform actions from valid 'words', which means they are guaranteed to carry out the required tasks.
This has nothing to do with communications between robots, and everything to do with finite-state machines being used to keep things in established states. The grammars are entirely internal to each robot's programming. There's a quote in the article from the research lead about machines programming themselves but that doesn't seem to have anything to do with the contents of the paper. The paper talks about followers/leaders but communication is extremely limited ("I'm not on a team yet" "okay I accept you to green team") and the "leaders" don't send any instructions in the grammar they designed.
Unless I'm missing a big chunk of the paper, the robots don't construct their own words at all, unless you mean "they have a short list of actions they can perform in different states and they pick from that list".
It's a neat study, and it's useful to explore the best design techniques for large scale swarms, but it sounds cooler and way different than it actually is