Abstract
<jats:p>The paper presents an integrated methodology for an AI-based hybrid framework designed to solve the continuous coverage problem under geometric constraints. The relevance of this work is driven by the wide range of applications of coverage models in telecommunications, logistics, robotics, territorial monitoring, and spatial planning, where classical discrete formulations become insufficient due to the complexity of real geographic regions and the necessity to optimize continuous parameters. The problem statement focuses on developing a methodological approach capable of combining complex geometric operations, global optimization procedures, and intelligent predictive models to enhance computational efficiency. The aim of the study is to construct a unified hybrid framework that integrates metaheuristic and memetic optimization methods with neural surrogate models and a UML-oriented information system architecture. The proposed methods include swarm and evolutionary algorithms, adaptive penalty mechanisms, neural approximation models, and combined techniques for coverage area evaluation. The results are presented in the form of a conceptual comparison of the efficiency of different framework components and their synergistic influence on the accuracy and performance of the optimization process. The conclusions emphasize the applicability of the integrated AI-based framework to large-scale coverage problems and its potential for further development towards dynamic and multi-criteria problem settings.</jats:p>