Evolutionary Robotics and Open-Ended Design Automation
Hod Lipson, Cornell University Can a computer ultimately augment or replace human invention? IMAGINE A LEGO SET AT YOUR DISPOSAL: Bricks, rods, wheels, motors, sensors and logic are your “atomic” building blocks, and you must find a way to put them together to achieve a given high-level functionality: A machine that can move itself, say. You know the physics of the individual components' behaviors; you know the repertoire of pieces available, and you know how they are allowed to connect. But how do you determine the combination that gives you the desired functionality? This is the problem of Synthesis. Although engineers practice it and teach it all the time, we do not have a formal model of how open-ended synthesis can be done automatically. Applications are numerous. This is the meta-problem of engineering: Design a machine that can design other machines. The example above is confined to electromechanics, but similar synthesis challenges occur in almost all engineering disciplines: Circuits, software, structures, robotics, control, and MEMS, to name a few. Are there fundamental properties of design synthesis that cut across engineering fields? Can a computer ultimately augment or replace human invention? While we may not know how to synthesize thing automatically, nature may give us some clues: After all, the fascinating products of nature were designed and fabricated autonomously.
In the last two centuries, engineering sciences have made remarkable progress in the ability to analyze and predict physical phenomena. We understand the governing equations of thermodynamics, elastics, fluid flow, and electromagnetics, to name but a few domains. Numerical methods such as finite elements allow us to solve these differential equations, with good approximation, for many practical situations. We can use these methods to investigate and explain observations, as well as to predict the behavior of products and systems long before they are ever physically realized. But progress in systematic synthesis has been much slower. For example, the systematic synthesis of a kinematic machine for a given purpose is a long-standing problem, and perhaps one of the earliest general synthesis problems to be posed. Robert Willis, a professor of natural and experimental philosophy at Cambridge, wrote in 1841 : [A rational approach to synthesis is needed] to obtain, by direct and certain methods, all the forms and arrangements that are applicable to the desired purpose. At present, questions of this kind can only be solved by that species of intuition that which long familiarity with the subject usually confers upon experienced persons, but which they are
totally unable to communicate to others. When the mind of a mechanician is occupied with the contrivance of a machine, he must wait until, in the midst of his meditations, some happy combination presents itself to his mind which may answer his purpose.” Robert Willis, Principles of Mechanism 
Almost two centuries later, a rational method for the synthesis in many domains is still not clear. Though many best-practice design methodologies exist, at the end of the day they rely on elusive human creativity. Product design is still taught today largely through apprenticeship: Engineering students learn about existing solutions and techniques for well-defined, relatively simple problems, and then – through practice – are expected to improve and combine these to create larger, more complex systems. How is this synthesis process done? We do not know, but we cloak it with the term “creativity”. The question of how synthesis of complex systems occurs has been divided in a dichotomy of two views: One view is that complex systems emerge through successive adaptations coupled with natural selection. This Darwinian process is well accepted in Biology, but is more controversial in engineering [2,35]. The alternative explanation...
References: 1. Alexander C. A., (1970) Notes on the Synthesis of Form, Harvard University Press 2. Basalla G., (1989) The Evolution of Technology, Cambridge University Press 3. Bentley, P. J. and Kumar, S. (1999). Three Ways to Grow Designs: A Comparison of Embryogenies for an Evolutionary Design Problem. Genetic and Evolutionary Computation Conference (GECCO '99), July 14-17, 1999, Orlando, Florida USA, pp.35-43. RN/99/2 4. Bongard J. C., Lipson H., (2004) “Once More Unto the Breach: Automated Tuning of Robot Simulation using an Inverse Evolutionary Algorithm”, Proceedings of the Ninth Int. Conference on Artificial Life (ALIFE IX), pp.57-62 5. Bongard J., Lipson H. (2004), “Automated Damage Diagnosis and Recovery for Remote Robotics”, IEEE International Conference on Robotics and Automation (ICRA04), pp. 3545-3550 6. Bongard, J. and Lipson, H. (2004) “Integrated Design, Deployment and Inference for Robot Ecologies”, Proceedings of Robosphere 2004, November 2004, NASA Ames Research Center, CA USA 7. Bongard, J. C. (2002) Evolved Sensor Fusion and Dissociation in an Embodied Agent, in Proceedings of the EPSRC/BBSRC International Workshop Biologically-Inspired Robotics: The Legacy of W. Grey Walter, pp. 102-109
8. Bongard, J. C. and R. Pfeifer (2003) Evolving Complete Agents Using Artificial Ontogeny, in Hara, F. and R. Pfeifer, (eds.), Morpho-functional Machines: The New Species (Designing Embodied Intelligence) Springer-Verlag, pp. 237-258 9. Bonner J.T., (1988) The Evolution of Complexity by Means of Natural Selection, Princeton University Press 10. Goldberg, D. E., (1989) Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley 11. Hartwell L.H., Hopfield J.H., Leibler S. and Murray A.W., 1999, “From molecular to modular cell biology”, Nature 402, pp. C47-C52 12. Hornby G.S., Lipson H., Pollack. J.B., 2003 “Generative Encodings for the Automated Design of Modular Physical Robots”, IEEE Transactions on Robotics and Automation, Vol. 19 No. 4, pp 703-719 13. Jakobi, N. (1997). Evolutionary robotics and the radical envelope of noise hypothesis. Adaptive Behavior, 6(1):131–174. 14. Koza J., (1992) Genetic Programming: On the Programming of Computers by Means of Natural Selection, MIT Press 15. Lipson H. (2004) "How to Draw a Straight Line Using a GP: Benchmarking Evolutionary Design Against 19th Century Kinematic Synthesis", Proceedings of Genetic and Evolutionary Computation Conference, Late Breaking Paper, GECCO 2004 16. Lipson H. (2004) "Principles of Modularity, Regularity, and Hierarchy for Scalable Systems", Genetic and Evolutionary Computation Conference (GECCO '04) Workshop on Modularity, regularity and Hierarchy 17. Lipson H., Pollack J. B. (2000) Automatic design and manufacture of artificial lifeforms. Nature, 406:974–978 18. Luke, S. and L. Spector. 1996. Evolving Graphs and Networks with Edge encoding: Preliminary Report. In Late Breaking Papers at the Genetic Programming 1996 Con-ference (GP96). J. Koza, ed. Stanford: Stanford Bookstore. 117-124 19. Malone E., Lipson H., (2004) “Functional Freeform Fabrication for Physical Artificial Life”, Ninth Int. Conference on Artificial Life (ALIFE IX), Proceedings of the Ninth Int. Conference on Artificial Life (ALIFE IX), pp.100-105 20. Melanie Mitchell, (1996) An introduction to genetic algorithms, MIT Press 21. Mytilinaios E., Marcus D., Desnoyer M., Lipson H., (2004) “Designed and Evolved Blueprints For Physical Self-Replicating Machines”, Proceedings of the Ninth Int. Conference on Artificial Life (ALIFE IX), pp.15-20 22. Nolfi S., Floreano D. (2004), Evolutionary Robotics: The Biology, Intelligence, and Technology of Self-Organizing Machines, Bradford Books 23. Papadimitriou C.H., Steiglitz K., Combinatorial Optimization : Algorithms and Complexity, Dover Publications
24. Parker, A.R., McPhedran, R.C., McKenzie, D.R., Botten, L.C. and Nicorovici, N.A.P., Aphrodite 's iridescence. Nature (2001) 409, 36-37 25. Paul, C. and J. C. Bongard (2001) “The Road Less Traveled: Morphology in the Optimization of Biped Robot Locomotion”, in Proceedings of The IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2001), Hawaii, USA 26. Preble S.F., Lipson H., Lipson M. (2004) "Novel two-dimensional photonic crystals designed by evolutionary algorithms", in M. Lipson, G. Barbastathis, A. K. Dutta, K. Asakawa (Eds.) Nanophotonics for Communication: Materials and Devices, Proceedings of SPIE Volume: 5597, pp. 118-128 27. Saylor J. Walker K., Moon F.C., Henderson D.W., Daimina D., Lipson H., Cornell University Digital Library of Kinematic Models (KMODDL), http://kmoddl.library.cornell.edu 28. Sims K. “Evolving 3D morphology and behaviour by competition”. Artificial Life IV, pages 28–39, 1994. 29. Suh N.P. 1990 The Principles of Design, Oxford University Press, Oxford UK 30. Wagner, G.P., Altenberg L., 1996 “Complex adaptations and the evolution of evolvability” Evolution 50:967-976 31. Watson, R.A. and Pollack, J.B. (1999). “Hierarchically-Consistent Test Problems for Genetic Algorithms”, Proceedings of 1999 Congress on Evolutionary Computation (CEC 99). Angeline, Michalewicz, Schoenauer, Yao, Zalzala, eds. IEEE Press, pp.1406-1413 32. Willis, R., (1841), Principles of Mechanism, London (available online at KMODDL ) 33. Wyatt D, Lipson H., (2003) “Finding Building Blocks Through Eigenstructure Adaptation”, Genetic and Evolutionary Computation Conference (GECCO ’03) 34. Yim, M., Zhang, Y. and Duff, D., "Modular Reconfigurable Robots, Machines that shift their shape to suit the task at hand," IEEE Spectrum Magazine cover article, Feb. 2002 35. Ziman J, (2003) Technological Innovation as an Evolutionary Process, Cambridge University Press 36. Zykov V., Bongard J., Lipson H., (2004) "Evolving Dynamic Gaits on a Physical Robot", Proceedings of Genetic and Evolutionary Computation Conference, Late Breaking Paper, GECCO '04
Please join StudyMode to read the full document