Stéphane MagnenatThe NewScientist article Robot swarms “evolve” effective communication said
Robots that artificially evolve ways to communicate with one another have been demonstrated by Swiss researchers. The experiments suggest that simulated evolution could be a useful tool for those designing of swarms of robots.
Roboticists Dario Floreano, Sara Mitri, and Stéphane Magnenat at the Swiss Federal Institute of Technology in Lausanne collaborated with biologist Laurent Keller from the University of Lausanne.
They first evolved colonies of robots in software then tested different strategies on real bots, called s-bots. Both simulated and real robots were set loose in an arena containing two types of objects one classified as “food” and another designated “poison” both lit up red.
Each bot had a built-in attraction to food and aversion to poison. They also have a randomly-generated set of parameters, dubbed “genomes” that define the way they move, process sensory information, and how they flash their own blue lights.
Stéphane Magnenat is a PhD student doing research
in mobile robotics. He loves the complexity of the universe,
the knowledge we have acquired
and structured as a species. He strongly believes that information
should be free, that our technologies
will reveal their true powers once people stop trying to “protect” and
reduce access to valuable information, such as computer program source
code, encyclopedias, teaching material, and artistic creations.
For those reasons, he’s very interested in Wikipedia and its
Stéphane coauthored Evolutionary Conditions for the Emergence of Communication in Robots, Division of labor and colony efficiency in social insects: effects of interactions between genetic architecture, colony kin structure and rate of perturbations, Superlinear Physical Performances in a SWARM-BOT, Emergence of signaling in colonies of simulated mobile robots, From S-bot to Swarmbot & Ishtar, an architecture for robotic control and monitoring, Ontogenèae (French), and De la nésessité des logiciels libres dans l’application de la démocratie en Suisse (French). Read his full list of publications!
He authored Enki, a fast 2D physics-based robot simulator. It is able to simulate cinematics, collisions, sensors and cameras of robots evolving on a flat surface. It also provides limited support for friction. It is able to simulate groups of robots hundred times faster than realtime on a modern desktop computer.
He coauthored Teem, a software framework that allows for easy creation and execution of evolutionary robotics experiments. Using a modular approach, it provides several genomes, evolution logics, neural networks topologies and neurons/synapses as well as sample experiments. It seamlessly integrates with the Enki simulator.
He coauthored Osqoop, an open source software oscilloscope. It features an arbitrary number of channels and long acquisition durations. Signal processing and external peripherals control is possible through a plugin architecture. Data sources are plugins as well.
He coauthored NanoC, a simple compiler for a subset of C for the PIC microcontroller. It takes an input in NanoC and produces CALM assembly source in output. He also coauthored PICsim, a simple PIC debugger. It can load an object file and the corresponding CALM assembly source code. Then it can set breakpoint, examine registers, do step by step. Its architecture is extensible for newer PIC. Read his full list of open source software!