|Paper title||Digital Twin: 3D Modelling and Simulation for Cities|
|Form of presentation||Poster|
We all agree since the 70’s that Earth Observation (EO) data is key to understand human activity and Earth changes. However two trends today are forcing us to rethink the use of EO to tackle new challenges:
- Georeferenced data sources, data quantity and quality keep increasing allowing a global and regular Earth coverage;
- AI and cloud storage allow swift fusion, analysis and dissemination of these data on online platforms.
Combined together, these two trends generate various reliable indicators. Once fused together, they will allow the anticipation of future humanitarian, social, economic and sanitary crisis, the adequate action plan to prevent them from happening, and, the provision of relief and support should they occur. Satellite imagery, 3D simulation, image analysis, mapping, georeferenced public and private data… we are getting enough tools to give the Earth a Digital Twin and this is not Science Fiction anymore.
Airbus and Dassault Systèmes joined forces to approach and reach this ambition focusing on cities. The imagery products from the Airbus constellation of satellites will be used with the simulation tools from Dassault Systèmes. The project aims at automatically building a 3D digital model of cities as well as simulating their entire environment and using them as baseline to digitise impactful events. It will cover the complete value chain of this 3D mapping analysis services: from data collection, 3D production models, simulation environment software ingestion, event and analysis models, to dissemination for studies. One of the specifics of the project is to tackle information from both a global perspective and small-scale details (0,1m, 0,05m…) in order to capture the impacts of urbanism on the environment, people’s health, economy and security.
To achieve a significant leap forward, the 3D environment used for the simulations will need to be more precise, be quickly generated, ready-to-use in a simulation environment, and made easily accessible to our customers, hence the following objectives:
• Increase the average location accuracy of the 3D models from 5-8m to 3-4m everywhere in the world, and allowing GPS reference points for use cases where submetric precision is required,
• Produce 3D models based on archive imagery (emancipating ourselves from the need to task satellites when time sensitiveness is high),
• Transform the current representation of 3D models (single canopy layer including ground, trees and man-made objects) to a model where we can isolate the ground and each 3D object,
• Ensure these 3D models can be ingested into the relevant simulation environments,
• Give rapid access to the 3D model database and the capacity to order them through our OneAtlas platform.
On the simulation side, the main challenges revolve around the need to develop new and robust methodologies for each use case (selection of relevant physical parameters, treatment of the 3D surface, hardware and software needs, overall quality expectation, etc.) and automate as many tasks as possible to reduce lead time.
There is a large variety of simulation domains such as aerodynamics, electromagnetism, hydrodynamics, fluid-structure interaction, passive scalar (pollutants, pathogens, radiologic threat agents…) or even energy performance in an urban area. Applications include construction and infrastructure industries, planning, energy, security and defence. And the simulation and results will be available via the 3DExpérience platform, as well as Airbus OneAtlas.
Such an approach combining high resolution EO and future proof simulation techniques takes is unique on the market and takes digital twins beyond the state of the art.