In this study, we present the implementation of invasive weed optimization (IWO) in the maximization of main-lobe to side-lobe level for the non-uniform planar antenna array. The antenna arrays investigated in this study are generated using the chaos game algorithm (CGA) and shaped into selected fractal geometries chosen on the basis of their interesting performance. This CGA is picked out in order to overcome the limitations found in the fractal arrays. All the attained results are compared with the results produced by a well-known optimization algorithm that is the particle swarm optimization (PSO). In all the optimized arrays, IWO shows superior optimization results compared with PSO.