Vision-Based Parameter Estimation of a Slung Load

This research paper introduces an innovative vision-based technique designed to locate and estimate the position of a tethered slung load relative to an Unmanned Aerial System (UAS). The approach relies on fundamental image processing methodologies, primarily emphasizing accurately determining the pitch of the slung load. This is achieved through a combination of Gaussian filtering, HSV filtering, and Fitzgibbon ellipse detection. An independent measuring device is employed to validate the calculated pitch, adding an extra layer of reliability to the vision software's output. The effectiveness of the proposed method is confirmed through rigorous indoor testing, utilizing measuring devices in controlled conditions. Additionally, outdoor scenarios showcase the reliability and feasibility of the vision-based approach. This approach holds significant promise for enhancing UAS operational capabilities, presenting a cost-effective vision solution for load positioning applications. The outcomes contribute substantially to the ad-vancement of UAS technologies, particularly in missions where precise load positioning is a critical determinant of success, extending the potential applications of vision-based systems in diverse operational environments.