Hardware evolution of analog circuits for in-situ robotic fault-recovery

We present a method for evolving and implementing artificial neural networks (ANNs) on field programmable analog arrays (FPAAs). These FPAAs offer the small size and low power usage desirable for space applications. We use two cascaded FPAAs to create a two layer ANN. Then, starting from a population of random settings for the network, we are able to evolve an effective controller for several different robot morphologies. We demonstrate the effectiveness of our method by evolving two types of ANN controllers: one for biped locomotion and one for restoration of mobility to a damaged quadruped. Both robots exhibit nonlinear properties, making them difficult to control. All candidate controllers are evaluated in hardware; no simulation is used.

[1]  Peter J. Bentley,et al.  An Evolutionary Approach to Damage Recovery of robot Motion with Muscles , 2003, ECAL.

[2]  Johannes Schemmel,et al.  Towards an artificial neural network framework , 2002, Proceedings 2002 NASA/DoD Conference on Evolvable Hardware.

[3]  Vu Duong,et al.  Evolution of analog circuits on field programmable transistor arrays , 2000, Proceedings. The Second NASA/DoD Workshop on Evolvable Hardware.

[4]  Ronald F. DeMara,et al.  Evolutionary fault recovery in a Virtex FPGA using a representation that incorporates routing , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[5]  Ricardo Tanscheit,et al.  Towards evolvable analog artificial neural networks controllers , 2004, Proceedings. 2004 NASA/DoD Conference on Evolvable Hardware, 2004..

[6]  Dario Floreano,et al.  Hardware spiking neural network with run-time reconfigurable connectivity in an autonomous robot , 2003, NASA/DoD Conference on Evolvable Hardware, 2003. Proceedings..

[7]  Garrison W. Greenwood,et al.  Fault recovery in linear systems via intrinsic evolution , 2004, Proceedings. 2004 NASA/DoD Conference on Evolvable Hardware, 2004..

[8]  Hod Lipson,et al.  Automated robot function recovery after unanticipated failure or environmental change using a minimum of hardware trials , 2004, Proceedings. 2004 NASA/DoD Conference on Evolvable Hardware, 2004..