Design of high-performance complex imaging systems enabled by automatic successive optimization

The traditional optical design process requires designers to intervene continuously from finding the starting point to obtaining the final design result, making it a time-consuming and labor-intensive task. At the same time, an increasing number of systems incorporate complex surface shapes and off-axis nonsymmetric structures, which help address imaging challenges but simultaneously introduce considerable design complexity. In this paper, we present an automatic design framework for high-performance complex imaging systems. By adopting a successive approach, an out-loop is introduced to enable fully automatic optimization. Backtracking and adjustment strategies are proposed to enable the timely correction of abnormal optimization and ensure the smoothness and correctness of the design process. The proposed framework enables transforming an all-plane starting point into a high-performance imaging system, which has long been a desired objective for optical designers. Diverse optimization routes can also be explored in order to get better design results. The feasibility of this design framework is validated by several design examples. This approach can significantly reduce manual intervention in the design process and minimize reliance on patents or existing starting points. Additionally, the design framework can be integrated into optical design software, running continuously in the background or on servers to automatically complete design tasks and find optimal results.