Evolution of Physical Machines

Can evolutionary principles be used to automatically design and manufacture physical machines and discover new concepts not foreseen by human engineers? Fully automated design and manufacture has long been the Holy Grail of the CAD/CAM field. In this research we explore some directions of this prospect. We evolve machines for specified tasks and replicate them into reality, thereby establishing for the first time a complete physical evolutionary design cycle into reality. To create a minimally biased design space, only two low-level building blocks are used: linear actuation elements and sigmoidal control neurons. These elements can assemble together to give rise to a universe of possible forms of controlled machines with rigid and flexible components. Quasi-static motion is then used to simulate low momentum behavior, including flexion, material failure, collision and friction. Selected machines are then automatically replicated into reality using rapid prototyping techniques. The automatically designed and manufactured machines then perform in reality. The paper discusses this approach and provides examples of both virtual and real machines evolved for the task of locomotion.

[1]  John Hallam,et al.  Evolving robot morphology , 1997, Proceedings of 1997 IEEE International Conference on Evolutionary Computation (ICEC '97).

[2]  Charles E. Taylor,et al.  Selection for wandering behavior in a small robot , 1995 .

[3]  I. Chen.  Theory and applications of modular reconfigurable robotic systems , 1994 .

[4]  Karl Sims,et al.  Evolving 3d morphology and behavior by competition , 1994 .

[5]  John J. Grefenstette,et al.  An Evolutionary Approach to Learning in Robots. , 1994 .

[6]  Hans P. Moravec.  Robot: Mere Machine to Transcendent Mind , 1998 .

[7]  Peter J. Bentley,et al.  Evolutionary Design By Computers , 1999 .

[8]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[9]  Wolfgang Banzhaf,et al.  Learning to Move a Robot with Random Morphology , 1998, EvoRobot.

[10]  Stefano Nolfi,et al.  Evolving non-trivial behaviors on real robots: A garbage collecting robot , 1997, Robotics Auton. Syst..

[11]  Gregory S. Chirikjian,et al.  Kinematics of a metamorphic robotic system , 1994, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.

[12]  Samir W. Mahfoud.  Niching methods for genetic algorithms , 1996 .

[13]  Phil Husbands,et al.  Evolutionary robotics , 2014, Evolutionary Intelligence.

[14]  Francesco Mondada,et al.  Mobile Robot Miniaturisation: A Tool for Investigation in Control Algorithms , 1993, ISER.

[15]  Toshio Fukuda,et al.  Genetic Evolution and Self-Organization of Cellular Robotic System , 1995 .

[16]  Pattie Maes,et al.  Increasing Adaptivity through Evolution Strategies , 1996 .

[17]  Maciej Komosinski,et al.  Framsticks: Towards a Simulation of a Nature-Like World, Creatures and Evolution , 1999, ECAL.

[18]  John Bares,et al.  Automated synthesis and optimization of robot configurations: an evolutionary approach , 1999 .

[19]  Stefano Nolfi,et al.  Evolving mobile robots in simulated and real environments , 1995 .

[20]  Philippe Bidaud,et al.  Genetic design of 3D modular manipulators , 1997, Proceedings of International Conference on Robotics and Automation.

[21]  Francesco Mondada,et al.  Automatic creation of an autonomous agent: genetic evolution of a neural-network driven robot , 1994 .

[22]  Frédéric Gruau,et al.  Cellular Encoding for interactive evolutionary robotics , 1996 .

[23]  Kazuyuki Murase,et al.  Genetic Evolution of a Logic Circuit which Controls an Autonomous Mobile Robot , 1996, ICES.

[24]  Randall D. Beer,et al.  Application of evolved locomotion controllers to a hexapod robot , 1996, Robotics Auton. Syst..

[25]  Pablo Funes.  Computer Evolution of Buildable Objects , 1997 .