A mutation operator for evolution strategies to handle constrained problems

In this paper we propose a new efficient mutation operator for evolution strategies (ES) the biased mutation operator (BMO). This operator is capable of improving the success rate to produce better offspring in constrained landscapes. The idea of our approach is to bias the mutation ellipsoid to lead the mutations into a more beneficial direction. Experimental results show that this bias enhances the solution quality in constrained search problems. The number of additional strategy parameters used in our approach equals to the dimensions of the problem. Compared with the correlated mutation, the BMO needs less memory. In addition, the BMO supersedes the computation of the rotation matrix of the correlated mutation and the asymmetric probability density function of the directed mutation. Therefore, it demands less computational cost and is easier to implement.