<p>Topology optimization adapted to additive manufacturing constraints has been an active area of research in recent years. However, the implementation of additive manufacturing restrictions in practical design cases often remains challenging due to convergence issues or suboptimal results when the constraints become difficult to satisfy. In this paper, a robust convergence algorithm is developed, incorporating parameter evolution strategies for the methods employed: density-based topology optimization, Heaviside filtering for achieving quasi-binary solutions, and Langelaar’s additive manufacturing filter. The algorithm is designed to provide robust convergence in cases where other methods fail, while maintaining high interpretability of the results for additive manufacturing. The algorithm is evaluated through a series of benchmark cases and compared against other available methods under challenging conditions. The results, in terms of solution interpretability (binarity) and robustness, are promising and are presented throughout this study.</p>