Martin Perris

Papers

Total Citations

H-Index

About

Martin Perris is a researcher in evolutionary robotics and artificial creativity, whose work explores how computational systems can autonomously develop physical behaviors. His most-cited paper, "Implicit Fitness Functions for Evolving a Drawing Robot" (2008), introduced a novel approach to robot evolution by replacing explicit performance metrics with implicit, emergent fitness criteria. This allowed a robotic arm to learn drawing behaviors without being directly programmed for artistic output, demonstrating how open-ended evolution can generate unexpected, creative solutions. With 13 citations, this foundational study has influenced subsequent research in embodied AI and evolutionary art. Perris’s contributions lie at the intersection of robotics, machine learning, and computational aesthetics, challenging conventional notions of both fitness functions and machine creativity. His work is particularly notable for its emphasis on implicit evaluation—a concept that has inspired further investigations into how robots can develop skills through interaction with their environment rather than predefined goals. For students and researchers, Perris’s research offers a compelling glimpse into how evolutionary principles can unlock novel behaviors in physical systems, bridging the gap between artificial life and artistic expression.