Emergence of communication and language in evolving robots

Abstract

Modeling the evolution of communication and language is one of the most fascinating and challenging problems in science. Progressing toward this ambitious goal implies explaining how simple communication forms emerged in the first place and how they evolved into structured communication systems with the characteristics of human language. In this chapter, we will illustrate how communication systems originate and evolve in a population of robots that adapt to a given task/environment. The analysis of these synthetic experiments can help us to understand: (i) how communication can emerge in a population of initially noncommunicating individuals, (ii) what conditions represent a prerequisite for the emergence of a robust and stable communication system, (iii) how the communication system changes by eventually increasing in complexity, and (iv) how signals and meanings originate and how they are grounded in robots' sensorimotor states.

improvement of signal communication for a foraging task using evolutionary robotics

“…In other words, robots associate the context in which a signal is produced and generate a significant message (Loula et al, 2010). Signals and their context are developed after exposition to adaptive conditions depending on environmental characteristics, which represent a factor that regulates the appearance (Nolfi, 2013) and establishment of robot communication skills (Montes-Gonzalez & Aldana-Franco, 2011).…”

Evolving goal-driven multi-agent communication: what, when, and to whom

“…Due to the complexity of human language [34], research efforts that model its evolution usually focus on only parts of the language, such as evolution of signals, evolution of symbols (symbolic communication), and evolution of words and syntax. The work in [6] explains the difference between words, signals, and symbols, and applies it to differentiate between models for the evolution of communication.…”