Drone imaging for industrial infrastructure inspection
This lecture overviews the use of drones for infrastructure inspection and maintenance. Various types of inspection, e.g., using visual cameras, LIDAR or thermal cameras are reviewed. Drone vision plays a pivotal role in drone perception/control for infrastructure inspection and maintenance, because: a) it enhances flight safety by drone localization/mapping, obstacle detection and emergency landing detection, and b) performs quality visual data acquisition. The drone should have: a) increased drone decisional autonomy and b) improved drone robustness and safety mechanisms (e.g., communication robustness/safety, embedded flight regulation compliance, obstacle avoidance). Therefore, it must be contextually aware and adaptive. Drone vision and machine learning play a very important role towards this end, covering the following topics: a) semantic world mapping b) drone and target localization, c) drone visual analysis for target/obstacle/point of interest detection, d) 2D/3D target tracking. Finally, embedded on-drone vision (e.g., tracking) and machine learning algorithms are extremely important, as they facilitate drone autonomy, e.g., in communication-denied environments. Two primary application areas are overviewed: a) industrial pipe damage detection and b) electric line/installation inspection. In the first application, pipe images are analyzed (segmented), damages are detected and are backprojected to the 3D pipe models and/or Lidar data. In the second application, electric line detection is performed and detection/tracking of electric towers and other entities (e.g., electric insulators).
The lecture will offer: a) an overview of all the above plus other related topics and will stress the related algorithmic aspects, such as: b) drone localization and world mapping, c) target detection d) target tracking and 3D localization and e) damage assessment. Some issues on embedded CNN, fast convolution and fast transformer computing will be overviewed as well.
Using Artificial Intelligence and Aerial Robots for defect detection and segmentation in Industrial Inspection of Viaduct
In this lecture, Begoña Arrue, of the GRVC Robotics lab Team, will present the use Artificial intelligence algorithms and tools for inspection analytics that can facilitate the analysis of the data gathered by the aerial robotic systems developed for Industrial Inspection of Viaduct in PILOTING project.
Lecture by Prof. Begoña Arrue.
Aerial Robotics for Industrial Inspection and the SIMAR project contribution
In this lecture, Miguel Ángel Trujillo, the EU SIMAR project coordinator, will present a few of the last European projects on aerial robotic inspection. Also, the talk will focus on how technology has evolved since the first one and the current state of the art. Furthermore, the SIMAR project will be presented, showing the consortium solution for the challenging use cases of pipe inspection for detecting corrosion under insulation.
Lecture by Dr Miguel Ángel Trujillo Soto.
Simulation of forest fires and floods
By using Unreal Engine for the creation of virtual environments we are able to simulate fire and flood scenarios with high accuracy and by using virtual UAVs we collect big virtual datasets. Mixing the virtual data with real-world data and training state-of-the-art machine learning models we hope to be able to detect real-world fires and floods with high precession.
Lecture by Evangelos Spatharis.
Flash flood modeling and in urban areas using High Resolution hydrodynamic model and machine learning models
Modeling flash floods in urban areas with complex topography is always challenging. Considering fine-scale hydrodynamic 2D shallow water model to perform simulations requires a lot of manual or semi-automatic data processing before being able to run simulations. This involves the transformation of high-resolution Digital Surface Model (Lidar) into a Digital Elevation Model that conserves the main hydraulic properties of the ground (culverts, weirs, barriers, etc) as well as accurate delineation of the streets and buildings, etc. In the context of the ExtremeXP project funded by the European Commission we assess the role of machine learning to improve the simulation and nowcasting (forecast with short term horizon) of flash flood events in the city of Nîmes in the South of France. First, we prepare all relevant datasets to design a fine scale 2D hydrodynamic model and then we calibrate it on several historical flood events. Once this model is calibrated and validated, we use it as a reference for conducting several scenarios of improvements using machine learning model. Two kinds of scenarios are analyzed. In the first kind lie all the machine learning techniques that would facilitate the design of the hydrodynamic model by either reducing the number of input data or reducing the necessary data transformation processes. The second kind of scenario consists in designing surrogates for the reference hydrodynamic model itself for nowcasting flood propagation during an event.
Lecture by K Larnier, J Coves, G Stephan and L. Dumas.
From Cloud to Edge through Microservices
Cloud Computing during the time has gained concrete evidence to be a disruptive technology still in its full development. Many drawbacks of the Cloud have brought to improve many their crucial aspects, like performance, security and privacy, etc. Today Edge Computing try to deal with these implications to make them less problematic and much more feasible. Starting from the NIST definition (IaaS, PaaS and SaaS), the talk looks at the last decade of ICT evolution preparing the systems for new ICT challenges and implementations, like AI algorithms on top of them.
Lecture by Prof. Massimo Villari.