Effect of Genetic Encoding on Evolution of Efficient Neural Controllers

摘要

In this paper, we present a new method based on multiobjective evolutionary algorithms to evolve low complexity neural controllers for the robots that have to perform two different tasks, simultaneously. In our method, each task and the structure of neural controller are considered as separated objective functions. We compare the results of two different encoding schemes: (1) Connectionist encoding and (2) Node based encoding. Simulation results show that multi-objective evolution can be successfully applied to generate low complexity neural controllers. In addition, node based encoding outperformed connectionist encoding in terms of robot performance and robustness of the neural controller.