Abstract
Localization has emerged as a critical problem over the past decades, with diverse techniques developed to address robot and mobile localization challenges across varied domains. However, existing localization methods still fall short of achieving the precision needed for certain high-demand applications. The proposed algorithm is designed to enhance localization accuracy by integrating mathematical modeling with the Crow Search Algorithm (CSA). The objective is to identify the most probable position within a designated search space. Anchored by a network of fixed points, the search area is initially defined. A mathematical approach is then applied to reduce this area by calculating the intersections between circles centered at each anchor point. Within this reduced area, an array of candidate points are selected, and their centroid is computed to serve as an initial estimate. The modified CSA iteratively improves upon this estimate by emulating the natural behavior of crows, updating its variables to converge on the optimal position. Experimental evaluations, conducted on both real and simulated datasets, demonstrate that the proposed algorithm leads to a better localization accuracy than existing methods. The proposed methodology achieves a significant accuracy improvement with an accuracy of 98%. These results confirm the effectiveness of our approach for applications that require high precision with minimal infrastructure and low computational complexity.