Effects of encodings and quality-diversity on evolving 2D virtual creatures

How to jointly optimize the morphology and controller is a challenging problem in evolutionary robotics. Due to the large search space, both quality diversity algorithms and types of encodings have been employed to search the solution space more effectively. Here we compare Multi-dimensional Archive of Phenotypic Elites (MAP-Elites) and a standard evolutionary algorithm as well as the effect of using a direct versus an indirect encoding. The results showed that the MAP-Elites algorithm found diverse solutions, yet the encodings accounted for a larger performance discrepancy. This indicates that the representation is at least as important as the optimization method for effectively creating robots.

[1]  Eric Medvet,et al.  Biodiversity in evolved voxel-based soft robots , 2021, GECCO.

[2]  Karine Miras,et al.  Constrained by Design: Influence of Genetic Encodings on Evolved Traits of Robots , 2021, Frontiers in Robotics and AI.

[3]  Kyrre Glette,et al.  MAP-Elites Enables Powerful Stepping Stones and Diversity for Modular Robotics , 2020, Frontiers in Robotics and AI.

[4]  Wojciech Matusik,et al.  RoboGrammar , 2020, ACM Trans. Graph..

[5]  Kyrre Glette,et al.  How Different Encodings Affect Performance and Diversification when Evolving the Morphology and Control of 2D Virtual Creatures , 2020, ALIFE.

[6]  Hod Lipson,et al.  Scalable co-optimization of morphology and control in embodied machines , 2017, Journal of The Royal Society Interface.

[7]  Yiannis Demiris,et al.  Quality and Diversity Optimization: A Unifying Modular Framework , 2017, IEEE Transactions on Evolutionary Computation.

[8]  Sebastian Risi,et al.  Evolution and Morphogenesis of Simulated Modular Robots: A Comparison Between a Direct and Generative Encoding , 2017, EvoApplications.

[9]  Takaya Arita,et al.  Artificial Metamorphosis: Evolutionary Design of Transforming, Soft-Bodied Robots , 2016, Artificial Life.

[10]  Antoine Cully,et al.  How do Different Encodings Influence the Performance of the MAP-Elites Algorithm? , 2016, GECCO.

[11]  Kenneth O. Stanley,et al.  Quality Diversity: A New Frontier for Evolutionary Computation , 2016, Front. Robot. AI.

[12]  Jean-Baptiste Mouret,et al.  Illuminating search spaces by mapping elites , 2015, ArXiv.

[13]  Hod Lipson,et al.  Unshackling evolution: evolving soft robots with multiple materials and a powerful generative encoding , 2013, GECCO '13.

[14]  Kenneth O. Stanley,et al.  Evolving a diversity of virtual creatures through novelty search and local competition , 2011, GECCO '11.

[15]  Hod Lipson,et al.  Generative representations for the automated design of modular physical robots , 2003, IEEE Trans. Robotics Autom..

[16]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[17]  Anthony Kulis,et al.  Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies , 2009, Scalable Comput. Pract. Exp..

[18]  A. E. Eiben,et al.  Introduction to Evolutionary Computing , 2003, Natural Computing Series.

[19]  R. Pfeifer,et al.  Evolving Complete Agents using Artificial Ontogeny , 2003 .