The value of information and the emergence of communication in evolved e-puck robots
Evolved e-puck robots communicate in proportion to the value of a one-bit food signal.
Simulated e-puck robots, reconstructed in PyBullet from a Webots model, evolve a one-bit directional signal while searching for a feeding zone detectable only when a robot passes over it. Controllers are GRU networks, alone or inside an Embedded Action Selection Architecture, and fitness is time spent over food. A compass to the food removes signaling entirely; without it, small-arena runs use a referential signal, a permanent beacon, or no signal, and the referential strategy generalizes best as the arena grows. Innate EASA reflexes reduce performance, and evolution consistently weakens them.
- Two simulated e-puck robots can share a one-bit directional signal.
- A food compass eliminates signaling; without it, strategies diverge by arena size.
- Referential signaling generalizes best and beats solitary search in large arenas.
- Evolving EASA priorities consistently weakens innate reflexes that hurt social performance.
From the original article
1 min read at huggingface.co
We study when communication emerges between autonomous agents and what role an innate action selection architecture plays. Two e-puck robots, reconstructed in PyBullet from their Webots model and validated against it, must find a feeding zone that can only be detected when a robot passes over it. Each robot can emit a one-bit signal whose direction is detected by the receiver.
By Fernando Montes-Gonzalez · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.