Our results show that for the single-leg locomotion problem, hypermutation increases the quality of the mrCGA’s solution in a dynamic environment, whereas the random immigrant variant produces slightly lower scores. Both of these variants can be easily added to the existing mrCGA hardware implementation without significantly increasing its complexity. In the future we plan to categorize the effects of the hypermutation and random immigrant strategies on the mrCGA for a variety of generalized benchmarks. This categorization will be useful to help determine which dynamic optimization strategy should be employed for a given problem.