Evolution of Synaptic Delay Based Neural Controllers for Implementing Central Pattern Generators in Hexapod Robotic Structures
We used synaptic delay based neural networks for implementing Central Pattern Generators (CPGs) for locomotion behaviors in hexapod robotic structures. These networks incorporate synaptic delays in their connections which allow greater time reasoning capabilities in the neural controllers, and additionally we incorporated the concept of the center-crossing condition in such networks to facilitate obtaining oscillation patterns for the robotic control. We compared the results against continuous time recurrent neural networks, one of the neural models most used as CPG, when proprioceptive information is used to provide fault tolerance for the required behavior.
Unable to display preview. Download preview PDF.
- 2.Campo, A., Santos, J.: Evolution of adaptive center-crossing continuous time recurrent neural networks for biped robot control. In: Proceedings European Symposium on Artificial Neural Networks (ESANN 2010), pp. 535–540 (2010)Google Scholar
- 5.Feoktistov, V.: Differential Evolution: In Search of Solutions. Springer, NY (2006)Google Scholar
- 7.Hexapod videos: http://www.dc.fi.udc.es/ai/~santos/Iwinac2015/Iwinac2015.html
- 9.Price, K.V., Storn, R.M., Lampinen, J.A.: Differential Evolution. A Practical Approach to Global Optimization. Natural Comp. Series. Springer (2005)Google Scholar
- 11.Santos, J.: Evolved center-crossing recurrent synaptic delay based neural networks for biped locomotion control. In: Proceedings IEEE Congress on Evolutionary Computation (IEEE-CEC 2013), pp. 142–148 (2013)Google Scholar
- 14.Smith, R.: Open Dynamics Engine (2003), http://opende.sourceforge.net