Abstract: In loitering missile-borne electro-optical reconnaissance systems, there is an inherent trade-off between detection range and coverage area of electro-optical sensors. Conventional approaches address this conflict by controlling sensor motion to achieve target scanning, thereby maintaining detection range while improving reconnaissance efficiency. However, existing reconnaissance scheme design methods primarily rely on empirical practices, suffer from strong dependence on designer experience, require multiple iterations of trial-and-error, and are prone to creating blind zones, making them inadequate for modern weapon systems. To overcome these challenges, an optimized ground-scanning reconnaissance scheme design method was proposed for loitering munition electro-optical systems. The method incorporated multiple influencing factors to construct an optimization model, including flight parameters, sensor performance parameters, servo system specifications and target dimensions. An optimization model for scanning reconnaissance schemes was established, with reconnaissance efficiency as optimization objective and target detectability, blind-zone-free coverage, and other design requirements as constraints. By employing an ant colony algorithm, the model yielded an optimal scanning reconnaissance scheme. Optimization results demonstrated that the proposed method significantly enhanced reconnaissance efficiency while fully satisfying all performance criteria.