Patch antenna design with Improved Sequential Quadratic Programming for automotive applications

This paper proposes a novel methodology to automate the design process of patch antennas using Improved Sequential Quadratic Programming (ISQP). Patch antennas are widely used in the automotive domain and are a key enabler of car-to-car and car-to-infrastructure communication. Although simulation tools have been improved in term of accuracy, a manual design process still takes a lot of time as well as computational efforts and relies to a large extent on human experience. In this paper, a fully automatic approach is presented and two major challenges are solved inherently. First, the conversion of a multiobjective problem into a single-objective problem is performed appropriately such that can be handled by classic Sequential Quadratic Programming (SQP). Second, the problem of being trapped in local optima is solved by a sophisticated exploration of the search space. Compared to other automation methods, our proposed methodology is fast, produces excellent results, needs limited computational efforts, no longer requires relevant domain knowledge, and is applicable to other types of antenna design. Experimental results of a truncated patch antenna with circular polarization are presented, achieving a good design point within 1470 seconds.