U-Flyte was launched in March 2018 in Waterford Airport by Maynooth University, SFI, and industry partners. The aim of the work is to tackle the current global bottleneck impeding the wider development of drone operations and roll-out of commercial services. The R&D work-plan is based around a series of inter-connected work-packages that deal with investigating, building and testing Unmanned Aircraft Systems (UAS) Traffic-Management (UTM), underpinned by a 3D drone airspace model (U-Space).
U-Flyte is a strategic research partnership, coordinated by Maynooth Universityand funded by Science Foundation Ireland, together with key industry collaborators including Airbus, Irelandia Aviation and INTEL.
The flight operations team operate the following drones:
- DJI Wind 4 – an industrial-level heavy lift quadcopter drone
- 3DR Solo – small, light and agile drone for camera operations
- Astec Falcon – high-performance quadcopter
- Quantum-Systems Trinity – hybrid VTOL/fixed-wing drone capable of longer flight
The U-Flyte ground station comprises a mobile trailer with power supply to allow the team to undertake field-based research and testing in remote areas around Ireland. We deployed the ground station during our Marine Watch demonstrator in Waterford in September 2018 to assist project partners and team members with various communications, flying and tracking tasks.
We operate a number of sophisticated cameras and sensors, including a Robin LiDAR and an Altum Micasense – a high resolution multispectral and thermal imaging camera
We run field tests at a number of sites around Ireland, including Waterford Airport, with the involvement of their ATC personnel.
Stephanie has a background in Geographical Information Systems, and obtained her PhD in climate change and the build environment at Maynooth University. She is the U-Flyte project manager.
Samuele has always loved computers and technology, and holds a bachelor's degree in Computer engineering and a Diploma in Machine Learning. He is currently working as a software developer on U-Flyte.
Rebekah has a Masters in GIS from Maynooth University and almost 10 years' experience designing and developing Geocomputational software and systems.
Aidan is an IRC/GeoAerospace PhD Candidate, focussing on Earth Observation, Drone Systems and Machine Learning
Sean is an SFI/Predict PhD Candidate, focussing on Earth Observation and Geocomputational Modelling
U-Flyte is currently recruiting PhD candidates to join this aerial robotics team based within the National Centre for Geocomputation (NCG), an established research centre at Maynooth University. We would especially like to hear from candidates with good primary degrees in Computing or Mathematics interested in developing careers in Aerial Robotics, Machine Learning, Computer Vision & Geospatial Data Analytics. If you feel you have the right background and want to know more – please see here.
- 3D Pathfinding & traffic management architectures for large-scale urban drone operations
This project focuses on the creation of new U-Space models & UTM systems designed to handle large numbers of autonomous drones carrying out a variety of data gathering, logistic and robotic activities in urban environments. This PhD will explore various algorithms and models that are required for constructing these new airspaces and enabling optimal traffic routing and overall management.
- Developing automated risk analysis and modelling tools for drone operations
Drone operation will always result in risk to varying degrees and both static and dynamic sources of risk need to be measured, classified and ultimately understood. Static risk includes vulnerable zones such as school yards, car-parks and exposed recreational parks. Dynamic risk includes weather and indeed other drones in flight. This PhD will deal with developing new methodologies and computational models to record, analyse and model risk as a fundamental input to U-Space/UTM system design.
- Sensor fusion for innovative Aerial and Ground based Detect and Avoid (DAA) systems
At the heart of this PhD will be the principle that drones can only fly safely, undertaking BVLOS operations, if all relevant aspects affecting flight safety are known. One key part of this is emerging optical, radar and RF-Sensor technology for detecting, locating and tracking both fixed and dynamic obstacles at low altitudes. This PhD will focus on how these new emerging technologies can be fused and analysed using the latest Machine Learning techniques to allow drones to fly autonomously over cities, towns and farms, along critical infrastructures and coastlines.
- Adaptable Machine Learning techniques for dynamic aerial scene understanding
Drones are capable of flying for more than an hour, covering tens of kilometres in distance, recording GigaBytes of combined optical and navigation data. Automated classification and measurement of man-made and natural features will become an increasingly important role for data-gathering drones. This PhD will investigate novel Machine learning (ML) tools and methodologies for identifying objects and activities in real-world scenes. These new ML tools will be tested and assessed for a variety of mapping, monitoring, defect inspection and anomaly detection tasks.