000155041 001__ 155041
000155041 005__ 20190619003305.0
000155041 0247_ $$2doi$$a10.5075/epfl-thesis-4941
000155041 02470 $$2urn$$aurn:nbn:ch:bel-epfl-thesis4941-1
000155041 02471 $$2nebis$$a6208227
000155041 037__ $$aTHESIS
000155041 041__ $$aeng
000155041 088__ $$a4941
000155041 245__ $$aGenetic Representation of Adaptive Neural Controllers
000155041 269__ $$a2011
000155041 260__ $$bEPFL$$c2011$$aLausanne
000155041 300__ $$a200
000155041 336__ $$aTheses
000155041 520__ $$aThe manual design of adaptive controllers for robotic  systems that face unpredictable environmental changes is  often challenging. There is thus a growing interest in the  development of automatic design tools to assist control  engineers. One of the most common approaches in this domain  is the evolutionary synthesis of Artificial Neural Networks  (ANNs), which allows the automatic design of control systems  with little or no human intervention. The performance of  evolved neural controllers, however, critically depends on  the choice of a suitable genetic representation, i.e.,  a description of the ANN on which the evolutionary search can  operate effectively. In this thesis, I propose an alternative to existing  representations for neural controllers based on an extension  of Analog Genetic Encoding (AGE), an abstraction of the  regulatory networks that control the expression of genes in  biological organisms. The proposed approach unites three  characteristics that have separately been considered  advantageous, but have so far not been aggregated into a  single representation: (1) it can be used to effectively  synthesize the topology, weights, and numerical parameters of  arbitrary networks; (2) it permits the use of heterogeneous  ANN models that feature multiple types of signals; and (3) it  facilitates the evolution of modular controllers. The results of a series of experiments demonstrate the  advantages of these characteristics: (i) neural architectures  synthesized with the proposed method display a performance  competitive to the best hand-designed networks and networks  synthesized with other representations; (ii) in tasks that  feature an unpredictable changing environment, controllers  with a heterogeneous ANN model outperform controllers with  conventional, homogeneous ANN models; and (iii) in tasks that  feature multiple modular sub-problems, the proposed  representation allows the automatic decomposition of the  problem, leading to a substantial improvement in the  performance of the evolved controllers. In summary, the proposed representation in combination  with a heterogeneous ANN model and an evolutionary algorithm  represents a promising method for the automatic synthesis of  adaptive control systems for robots.
000155041 6531_ $$aEvolutionary Robotics
000155041 6531_ $$aEvolutionary Algorithms
000155041 6531_ $$aArtificial Neural Networks
000155041 6531_ $$aGenetic Representation
000155041 700__ $$0243223$$g167254$$aDürr, Peter
000155041 720_2 $$aFloreano, Dario$$edir.$$g111729$$0240742
000155041 8564_ $$uhttps://infoscience.epfl.ch/record/155041/files/EPFL_TH4941.pdf$$zTexte intégral / Full text$$s5012142$$yTexte intégral / Full text
000155041 909C0 $$xU10370$$0252161$$pLIS
000155041 909CO $$pthesis-bn2018$$pthesis-public$$pDOI$$ooai:infoscience.tind.io:155041$$qGLOBAL_SET$$pSTI$$pthesis$$qDOI2
000155041 918__ $$dEDIC2005-2015$$cIMT$$aSTI
000155041 919__ $$aLIS
000155041 920__ $$b2011
000155041 970__ $$a4941/THESES
000155041 973__ $$sPUBLISHED$$aEPFL
000155041 980__ $$aTHESIS