in this article is part of the ongoing activities performed in the framework of the EASY-PV project sanctioned under H2020Galileo-2015-1 call. THE PAYLOAD DESCRIPTION AND SPECIFICATIONS heart of the payload is represented by the OBC element that has in charge different tasks such as: C C In Figure 1, the architecture of the entire Easy-PV RPAS is represented. The payload designed for this specific application consists in: handling sensors acquisitions; handling commands for payload control from pilot or payload operator; C storing thermal and optical acquired images; C hosting and handling computer vision algorithm; C a thermal camera; C hosting and handling the GEO-Tagging SW module; C an optical camera and a gimbal; C managing acquired data towards the Service Center. C an On-Board Computer (OBC). The most important aspect in the payload designing is the realization of the specific InterFace block required to command and control the thermal sensor and all the mechanism necessary to keep the pointing error as low as possible and to keep the device aligned for optimal nadiral acquisition during the RPAS movements. The Figure 1. Easy-PV RPAS Architecture model. AUGUST 2018 In Figure 2, a block diagram of the implemented payload is highlighted. The blue blocks represent the hardware modules of the payload; the yellow blocks represent the main software components. The acquired optical and thermal images are geo-tagged with a precise position information obtained from the GNSS C94-M8P Ublox receiver. The position (referred to the center of the antenna) is shifted into the thermal camera optical center. Such information is used by the OBC element to tag each image center with accurate geographical references. Moreover, the computer vision block has embedded PV panel shape recognition features, while the typologies of thermal anomalies are handled as plug in. Such information is available to the pilot (or payload operator) as screen overlay in real time to add additional remarks to each panel (identified or not) in the case of need. For instance, in the case of bad identification of PV panels or the case of false positives from the algorithm, the pilot (or payload operator) may flag suspect panels during flight operations through his remote control screen. Such information is useful in after-flight post-processing to allow IEEE A&E SYSTEMS MAGAZINE 59