Artificial Metamorphosis: Evolutionary Design of Transforming, Soft-Bodied Robots

被引:30
作者
Joachimczak, Michal [1 ]
Suzuki, Reiji [1 ]
Arita, Takaya [1 ]
机构
[1] Nagoya Univ, Dept Complex Syst Sci, Grad Sch Informat Sci, Chikusa Ku, Furo Cho, Nagoya, Aichi 4648601, Japan
基金
日本学术振兴会;
关键词
Metamorphosis; artificial development; soft robotics; body-brain coevolution; automated design; evolutionary algorithm; artificial life;
D O I
10.1162/ARTL_a_00207
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We show how the concept of metamorphosis, together with a biologically inspired model of multicellular development, can be used to evolve soft-bodied robots that are adapted to two very different tasks, such as being able to move in an aquatic and in a terrestrial environment. Each evolved solution defines two pairs of morphologies and controllers, together with a process of transforming one pair into the other. Animats develop from a single cell and grow through cellular divisions and deaths until they reach an initial larval form adapted to a first environment. To obtain the adult form adapted to a second environment, the larva undergoes metamorphosis, during which new cells are added or removed and its controller is modified. Importantly, our approach assumes nothing about what morphologies or methods of locomotion are preferred. Instead, it successfully searches the vast space of possible designs and comes up with complex, surprising, lifelike solutions that are reminiscent of amphibian metamorphosis. We analyze obtained solutions and investigate whether the morphological changes during metamorphosis are indeed adaptive. We then compare the effectiveness of three different types of selective pressures used to evolve metamorphic individuals. Finally, we investigate potential advantages of using metamorphosis to automatically produce soft-bodied designs by comparing the performance of metamorphic individuals with their specialized counterparts and designs that are robust to both environments.
引用
收藏
页码:271 / 298
页数:28
相关论文
共 41 条
[1]  
[Anonymous], 1995, THESIS
[2]  
[Anonymous], P INT C SIM SYNTH LI, DOI DOI 10.7551/978-0-262-31050-5-CH056
[3]  
[Anonymous], 2011, PRINCIPLES DEV
[4]  
Bongard JC, 2003, MORPHO-FUNCTIONAL MACHINES: THE NEW SPECIES, P237
[5]   Morphological change in machines accelerates the evolution of robust behavior [J].
Bongard, Josh .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2011, 108 (04) :1234-1239
[6]   An octopus-bioinspired solution to movement and manipulation for soft robots [J].
Calisti, M. ;
Giorelli, M. ;
Levy, G. ;
Mazzolai, B. ;
Hochner, B. ;
Laschi, C. ;
Dario, P. .
BIOINSPIRATION & BIOMIMETICS, 2011, 6 (03)
[7]  
Carroll SB, 2005, ENDLESS FORMS MOST B
[8]  
Cheney N, 2013, GECCO'13: PROCEEDINGS OF THE 2013 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, P167
[9]   Novelty-Based Evolutionary Design of Morphing Underwater Robots [J].
Corucci, Francesco ;
Calisti, Marcello ;
Hauser, Helmut ;
Laschi, Cecilia .
GECCO'15: PROCEEDINGS OF THE 2015 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, 2015, :145-152
[10]  
Deb K., 2009, MULTIOBJECTIVE OPTIM