This paper addresses the problem of designing optimal satellite constellations for Intensity Correlation Interferometry (ICI) missions based on the intensity correlation approach. Optimality in ICI missions requires maximizing the coverage of the frequency resolution disc. The 3-D Lattice Flower Constellations Theory, which includes the J
2gravitational perturbation, is here applied to optimize the constellation design. Optimization is performed using a Genetic Algorithm with two types of fitness functions to estimate the optimal design parameters. The fitness function describing the design optimality is constrained by altitude range to avoid perigees that are too low (drag) and apogees that are too high (inner Van Allen belt). Numerical results clearly show the resolution disc coverage as a function of the number of satellites and the frequency discretization bin number (pixels). Performances and results are provided to help select a fitness function for the mission and constellation designer.