Periodic Reporting for period 1 - REFMAP (Reducing Environmental Footprint through transformative Multi-scale Aviation Planning)
Reporting period: 2023-02-01 to 2024-07-31
REFMAP aims to develop a digital service for quantifying and minimizing the environmental footprint of air mobility at a “multi-scale” level. The core objective is to create an online platform through widespread collaboration with international partners to optimize both individual flight trajectories (micro-scale), such as unmanned aircraft systems (UAS), and overall air traffic (macro-scale). The research conducted in REFMAP will investigate a new aviation business model and how this model is affected by environmental data (e.g. weather, local pollution, wind intensity), emissions from different types of air vehicles (such as CO2 emissions or noise), and their trajectories.
The analytics platform will implement numerical simulations, predictive models, and deep-learning methods. The latter will help to (i) predict the optimal trajectories given a series of environmental data and (ii) develop a business model aligned with the EU’s Green Agenda.
The overall objectives of REFMAP are:
1). To produce robust real-time artificial intelligence (AI) models that optimize environmental performance in multi-scale air traffic management (ATM), including data exchanges with U-space. These models will be trained using multi-fidelity flow data;
2). To develop a framework that facilitates the responsible, sustainable, interoperable, coordinated, and safe expansion of the UAS industry;
3). To model the noise of commercial aviation and drones (as novel air technology) and define targets for trajectory optimization to minimize the impact on exposed communities and wildlife, ensuring a quieter, greener, and more sustainable aviation sector;
4). To develop innovative environmental key performance indicators (KPIs) related to multi-scale European aviation business;
5). To validate the newly developed models, services, and KPIs through large-scale simulations to assess their impact on existing ATM and emerging U-space services;
6). To align its functionalities and capabilities with the needs of aviation stakeholders to accelerate deployment and enable new aviation business models.
The analytics platform will implement numerical simulations, predictive models, and deep-learning methods. The latter will help to (i) predict the optimal trajectories given a series of environmental data and (ii) develop a business model aligned with the EU’s Green Agenda.
The overall objectives of REFMAP are:
1). To produce robust real-time artificial intelligence (AI) models that optimize environmental performance in multi-scale air traffic management (ATM), including data exchanges with U-space. These models will be trained using multi-fidelity flow data;
2). To develop a framework that facilitates the responsible, sustainable, interoperable, coordinated, and safe expansion of the UAS industry;
3). To model the noise of commercial aviation and drones (as novel air technology) and define targets for trajectory optimization to minimize the impact on exposed communities and wildlife, ensuring a quieter, greener, and more sustainable aviation sector;
4). To develop innovative environmental key performance indicators (KPIs) related to multi-scale European aviation business;
5). To validate the newly developed models, services, and KPIs through large-scale simulations to assess their impact on existing ATM and emerging U-space services;
6). To align its functionalities and capabilities with the needs of aviation stakeholders to accelerate deployment and enable new aviation business models.
The project performed a series of propaedeutic activities for the overall output of the project such as the REFMAP online platform and a new concept of business model for European aviation.
The main tasks completes from the project regards:
1). high-fidelity simulations,
2). the creation and validation of an air-quality model,
3). a new ML-based reconstruction framework to reconstruct the flow field in urban-like environment in real time,
4). experiments to evaluate the noise impact on humans and wildlife,
5). the training of the DRL model for trajectory optimisation in 2D,
6). the integration of the flight dynamics models in a pre-existing framework,
7). the creation of interview and the definition of the Minimum Viable Product, necessary for the business model
The main tasks completes from the project regards:
1). high-fidelity simulations,
2). the creation and validation of an air-quality model,
3). a new ML-based reconstruction framework to reconstruct the flow field in urban-like environment in real time,
4). experiments to evaluate the noise impact on humans and wildlife,
5). the training of the DRL model for trajectory optimisation in 2D,
6). the integration of the flight dynamics models in a pre-existing framework,
7). the creation of interview and the definition of the Minimum Viable Product, necessary for the business model
During the first period, we focused on modeling the environmental impact of Urban Air Mobility (UAM) using numerical simulations and experimental data. This impact was classified into two main areas: the effects of combustion and noise. This analysis aims to support the development of new European business models and products (e.g. Urban Air Mobility (UAM), seaplanes) with minimal environmental impact, enhancing European competitiveness.
We also performed a series of experiments to evaluate the noise annoyance on human health. This is the first attempt to evaluate the impact of the noise emitted by drones and commercial aviation on the European citizens and the wildlife. This would be achieved through the combination of numerical simulations and the modelling of the environmental impacts and the fuel emissions.
We also promoted the use of the artificial intelligence for decision-making in trajectory optimization by demonstrating that it can be used to reduce the operating cost.
The prediction of the wind pattern can also used to prevent the cities from extreme meteorological events.
All the aforementioned novelties from our project would reduce the overall environmental impact of transport.
We also performed a series of experiments to evaluate the noise annoyance on human health. This is the first attempt to evaluate the impact of the noise emitted by drones and commercial aviation on the European citizens and the wildlife. This would be achieved through the combination of numerical simulations and the modelling of the environmental impacts and the fuel emissions.
We also promoted the use of the artificial intelligence for decision-making in trajectory optimization by demonstrating that it can be used to reduce the operating cost.
The prediction of the wind pattern can also used to prevent the cities from extreme meteorological events.
All the aforementioned novelties from our project would reduce the overall environmental impact of transport.