4IPLAY - Improving Intelligent Infrastructure Inspection by Pushing UAS Level of Autonomy in Challenging Environments

The 4IPLAY project wants to develop innovative so- lution to demonstrate in relevant environments the concept of re- peatable automated visual inspection of infrastructures as bridges based on autonomous drones. 4IPLAY builds on the know-how experience of a multi-disciplinary research team and is focus on three main pillars: (i) automated acquisition and processing of multi-spectral data using artificial intelligence approaches; (ii) advanced flight control logics which are tailored for the conditions encountered in the proximity of large infrastructures; (iii) distributed multi-drone architectures to support real time cooperative navigation and off-line trajectory reconstruction. The analysis of images using artificial intelligence approaches will be one of the pillar, with a focus on the detection of defects also using generative techniques to augments dataset. Two key aspects are analyzed for the control system: the definition of a data-driven model for ceiling and wall effects and a disturbance-rejection control system for an Unmanned Aerial Vehicle. The research on cooperative multi-drone systems includes 1-to-N planning and control, as well as tight integration of cooperative navigation and distributed sensing. The paper introduces the main research areas of the project and provides an overview of current progress.