Evolving control strategies for suppressing heteroclinic bursting

We combine a detailed understanding of the dynamics of low-dimensional models of bursting in the turbulent boundary layer with the method of genetic programming to obtain appropriate control strategies for the suppression of such bursting in these models. The study is applicable to O(2) symmetric systems for which structurally stable heteroclinic cycles are the dominant dynamical features. We argue that such a combined approach can prove a useful tool in achieving control in higher-dimensional models where actual analysis is prohibitively complicated. The results of the present study are compared to near-optimal controllers derived in previous studies. (Author)