This academic article explores the concept of Morpho-Evolution (ME), which is the simultaneous optimization of a robot’s design and controller to maximize its performance. The researchers use Local Optima Network (LON) analysis to investigate the fitness landscapes induced by different encodings when evolving a robot for a locomotion task. The findings from this analysis will facilitate the design of new algorithms or operators that are customized to ME landscapes in the future.

 

Publication date: 13 Feb 2024
Project Page: https://doi.org/10.1145/nnnnnnn.nnnnnnn
Paper: https://arxiv.org/pdf/2402.07822

Post Views: 219