As part of its InCubed co-funded MSCM initiative, Aerospacelab is currently looking for input from users of Sentinel-2 products in order to respond to market needs and improve downstream solutions.
The Multispectral Companion Mission (MSCM) is a constellation currently under development and aimed at supplementing Copernicus Earth observation products with small multispectral satellites, giving rise to high-quality, daily-coverage data sets which are strongly complementary to Sentinel-2 data. Belgian company Aerospacelab is developing and deploying an In-Orbit Demonstrator (IOD) for MSCM in an activity launched in February of this year and supported by ESA’s InCubed programme. The new data products emerging from MSCM, mainly characterised by a high temporal sampling rate, will allow current Sentinel-2-based models to be enhanced and will also unlock new use cases in the fields of precision agriculture, food security and environmental monitoring.
To ensure that its product planning is suitably matched to the needs of the user community, Aerospacelab is seeking collaboration from Sentinel-2 users. Participating entities will be asked to complete short surveys and tests covering topics such as image quality requirements and data access. The respondents will then have access to the resulting report, which will summarise current end-user market trends.
Taking part in this market information exercise will give organisations the opportunity to shape the path of the development activity, as the MSCM constellation will be tailored to the input received from key users. Participants will also have free access to data acquired during the Beta in-orbit testing phase.
In three key sessions today, the Φ-week event dealt with the real nitty-gritty of advanced computational methods in Earth observation. Contributors discussed Artificial Intelligence for Earth Observation and Science (AI4EO and AI4SCIENCE), along with the emerging potential of quantum computing.
This morning’s session was entitled ‘AI4EO – Learning from Earth Observation Data to Understand Our Planet’. In summing up the role of Artificial Intelligence (AI), session co-chair Pierre Philippe Mathieu (Head of the Φ-lab Explore Office) gave a simple definition: “When AI meets Big Data from space – that’s where the magic happens.” He went on to describe just why AI is becoming such an essential tool: “Classical methods for information retrieval are now reaching their limits as a result of the volume and variety of data generated. AI offers a completely new view of acquired data, but also automates and accelerates data insight. One of the main aims of the session is to illustrate how to deliver the maximum value from the marriage of AI and EO data, while highlighting the limitations of the technology.”
Adopting AI methods is not without its challenges, and Φ-lab visiting professor Begüm Demir kicked off the session with a talk on the issues surrounding label noise in the training of Deep Learning algorithms. The presentation included proposals for identifying and reducing the impact of noisy image labels in land-cover annotations. The question of uncertainty was also taken up in talks on Machine Learning (ML) in climate change prediction and spatial mapping, while human development featured strongly in presentations on slum improvements, urban settlement and food security.
A feature of AI4EO which is central to the aims of the Φ-sat-1 experiment is filtering data directly on the satellite. A talk from Frontier Development Lab presented research on ML-driven unsupervised novelty detection which has, as Postdoctoral Research Fellow Valentina Zantedeschi from INRIA and University College London explained, already produced some promising results: “We’ve shown that unsupervised ML techniques improve the detection of relevant change while being less susceptible to noise than standard differencing methods. They also promise to reduce compute and memory requirements in terms of data processing and storage.”
“The broad range of talks has given an invaluable insight into the possibilities and hurdles associated with AI in EO,” commented the session’s other chair, ESA Research Fellow Rochelle Schneider. “I was particularly impressed by the diversity of our speakers and their shared passion for both innovation and pushing the boundaries of EO information processing.”
First up after lunch was the related topic of AI4SCIENCE. The session was chaired by ESA Open Science Platform Engineer Anca Anghelea: “AI is now showing enormous potential in helping us answer some fundamental geoscience questions. Machine Learning in particular can play a major role in rapid knowledge discovery by learning patterns and models from data.” With contributions from NASA and Microsoft as well as academia, the talks covered explainable AI and causal discovery, big Earth systems science, sustainability and seasonal forecasting.
Recent advances in quantum computing are expected to unleash unprecedented computing power in the near future for processing and analysing EO data, for example from Synthetic-Aperture Radar (SAR), multispectral and hyperspectral imaging. Following the launch of ESA QC4EO at last year’s Φ-week, the final session of the group showcased some of the work carried out in this field in the last twelve months. CERN presented both the status of its research in quantum computing and the associated shared experiences for EO and particle physics. Other institutions taking part included ECMWF and Oxford University, with the latter’s Timothy Palmer presenting the imaginatively titled ‘Quantum Computing for Earth Observations: the Good, the Bad and the Noisy’.
“Quantum computing and artificial intelligence are new digital technologies that can be combined to get better EO products, and get them faster,” said ESA Digital Technologies Engineer and session facilitator Bertrand Le Saux. “We need to develop and harness the power of Quantum Machine Learning for EO Big Data analytics and, as today’s speakers have shown, we’re now establishing the groundwork for exploring how the unique potency of quantum computing can be used for extracting information from EO data.”
How can a digital replica of Earth help us understand our planet’s past, present and future? As part of the fourth edition of Φ-week taking place this week, a group of European scientists have put forward their ideas on the practical implementation of Digital Twins and the potential application areas for a Digital Twin Earth in the real world.
In the coming decades, population growth and human activities are expected to amplify the current pressures on critical resources such as fresh water and food, intensify the stress on land and marine ecosystems, as well as increase environmental pollution and its impacts on health and biodiversity.
These threats, comprising rising sea levels, increasing ocean acidification and more intense extreme events like floods and heatwaves, will need to be closely monitored, especially for our most vulnerable populations.
Cloud-based EO platforms are an integral element for facilitating the paradigm change in Earth Observation (“bringing the algorithm to the data” et al.). Digital EO platforms are “Enablers” as they can boost the productivity of scientific/commercial users by simplifying the commonly encountered complex EO workflows. Key enabling elements include for example powerful and intuitive analytical operators, scalability of processing operations, efficient pixel-level data access and dynamic allocation of compute resources. The “Enabling EO Platforms” session at the 2021 ESA Φ-Week will showcase some key activities in the EO platforms domain.
First, openEO platform will be introduced by Alexander Jacob (EURAC). It is a highly federated, open-source EO analytics environment that abstracts underlying complexities when extracting value-added information from petabyte-scale data archives. The launch as an operational service will take place during this Φ-week with core deployments in multiple cloud environments (EODC/VSC4, CreoDIAS, TerraScope plus a data federation with SentinelHub/EuroDataCube).
With a boom in the global market for Earth observation information and data products, participants at this year’s Φ-week conference have been digging deep into the ‘market perspective’. This important topic includes how to gain a better understanding of what governments, industry, the public and other users of Earth observation products and services need and expect to create value for society and the economy.
According to the latest report from the Geospatial Industry Outlook and Readiness Index, GeoBuiz, the Earth observation industry as a whole is estimated to have been worth almost US $58 billion in 2019, rising to almost US $76 billion in 2020. Moreover, it is thought that this steep rise merely marks the beginning of bigger things to come.
In addition, with respect to Europe industry, the 2021 survey published by the European Association of Remote Sensing Companies, EARSC, indicates that in 2020 there was a 24% growth in the number of companies associated with the Earth observation industry, as well as a 24% increase in revenue and a 17% rise in employees, compared to 2019.
Two key sessions on today’s Φ-week agenda painted a bright future for the proliferation of AI-backed Earth observation applications. With the latest news on the Φ-sat missions, together with examples showcasing the widening array of commercial and institutional use cases, the speakers presented a wealth of inspiring and thought-provoking material.
This morning’s session was devoted to the Φ-sat concept, covering both the achievements of the Φ-sat-1 experiment and the development of its next incarnation, Φ-sat-2. Launched just over a year ago, the Φ-sat-1 experiment was set up to study the impact of onboard Artificial Intelligence (AI) on EO data collection. The project developers Cosine Remote Sensing and the University of Pisa were on hand during the session to report on progress. Spain’s Universitat Politècnica de Catalunya was also well represented, with four talks on subjects such as soil moisture and sea-ice extent from the FSSCat cubesat-based mission, which is also hosting the Φ-sat-1 experiment.
Other Φ-sat-1 suppliers also took part, including Ubotica Technologies. The company presented results on a topic at the very heart of the Φ-sat-1 experiment mission: the Movidius Myriad 2 AI Accelerator. This chip is the engine behind the satellite’s data processing capabilities and, as Ubotica’s Chief Technology Officer Aubrey Dunne explained, its performance is being monitored continuously while in-flight: “Φ-sat-1 has provided unique in-flight performance data for the Myriad 2 AI accelerator, giving valuable insight into the reliability of the device and the accuracy of its AI computations. Results of in-flight self-test diagnostics from Φ-sat-1 indicate that Myriad 2 was 100% functional and performed all verifiable inference requests flawlessly, demonstrating the reliability and practicality of AI on cubesats and providing valuable flight heritage to support its integration into future AI-rich satellite missions.”
Φ-sat-2 is being developed by a group of entities across Europe. Lead company Open Cosmos was at the session to give an overview of the platform architecture and its key features. “The biggest challenge on this mission for us is facilitating all the needs of the AI applications that will be running onboard,” said Irina Babkina, Systems Team Lead for the Φ-sat-2 project at Open Cosmos. “What we’re working on at the moment to meet that challenge is ensuring that the application-enabling algorithms we’ll be running actually have the required performance.”
At the helm during the presentations and round-table discussion was ESA Φ-sat activity lead Massimiliano Pastena. He was impressed by the progress made on both mission variants: “Φ-sat-1 is generating useful data which is helping to demonstrate the worth of onboard AI in EO satellites. This is laying the foundation for the parallel development of Φ-sat-2, which is scheduled for launch next year. As we’ve heard today, the power to upload, deploy and update AI applications in orbit, almost as simply as we do on our smartphones, will ensure that Φ-sat-2 becomes the benchmark for the next generation of flexible, user-oriented nanosatellites.”
The theme of targeting user needs continued in the afternoon with the System of Systems session. Giuseppe Ottavianelli, Head of ESA’s Earth Observation Applications Section and one of the session chairs, outlined the scope of the gathering: “The system-of-systems concept is about the possibilities deriving from the synergy between publicly-funded EO missions and private commercial assets. With global coverage and the long-term availability of calibrated data, public missions such as the Copernicus Sentinels form a solid basis for commercial investments in terms of both the space segment and downstream value-adding industry, giving rise to a host of customer-tailored applications.”
The session provided a fitting exposition of the breadth of such applications. Alongside the more established areas of agriculture, forestry and coastal monitoring, the speakers also addressed infrastructure asset management, urban planning and even the use of Very High Resolution (VHR) data for security applications. One talk in particular, given by ESA CHIME Operations Manager Antonio Gabriele, highlighted the potential gains from public-private EO initiatives and explored the idea of a hybrid constellation of CHIME and VHR smallsats, along with the applications that could result from this combined fleet.
Commenting on this and the other talks in the session, Head of the ESA EO System Architect Office Antonio Ciccolella was upbeat about the prospects for continuing growth in the sector: “The CHIME+VHR hybrid idea shows just how the system-of-systems concept could bring significant benefits for society at large. Underpinned by technologies such as AI, the Internet of Things and Blockchains, we expect the emerging abundance of new hardware and applications to deliver what could be called the Future Earth Intelligence System. The presentation from Rafael Guzman of Satlantis is another concrete example of private-sector investment in space, complementing Copernicus observations and using Sentinel data as a reference for the endeavour.”
ESA InCubed and the German Research Center for Artificial Intelligence (DFKI) have signed a contract to support a new development initiative focused on Artificial Intelligence for Earth Observation (AI4EO). The three-year programme will involve setting up an innovation seedbed where DFKI’s AI experts will collaborate with major industry players on new EO business cases.
Earth observation is enjoying an ever-widening field of applications, from crop yield forecasting and land cover detection to urban planning and disaster management. However, with the sheer volume of data generated, simple manual analysis is often no longer practicable, giving rise to the need for automatic Artificial Intelligence (AI) tools such as Machine Learning (ML).
The funding contract for the AI4EO Solution Factory was confirmed yesterday at this year’s ESA Φ-week event. Commenting at the signing ceremony, ESA Acting Director for Earth Observation Programmes Toni Tolker-Nielsen emphasised the landmark nature of the activity: “I’m very pleased to be here for the launch of this initiative, which is the first of its kind supported by ESA. We see a great deal of potential in the AI4EO Solution Factory, and I look forward to following its progress in delivering AI-powered EO solutions to customers.”
“This exciting InCubed activity will allow DFKI, together with major German industry partners, to show what great benefits Earth observation can unfold through the use of AI. We anticipate that the AI4EO Solution Factory will break new ground in the commercialisation of Earth observation for various industry sectors,” added Michael Nyenhuis, InCubed Programme Coordinator at the German Space Agency at DLR.
Prof. Andreas Dengel, Executive Director at DFKI Kaiserslautern and Head of Smart Data & Knowledge Services, spoke of some of the envisaged benefits: “The AI4EO Solution Factory creates a collaborative environment to open up new business scenarios in the field of Earth observation, based on the combined expertise of DFKI and ESA, and to develop customised AI solutions for partners and users. Our transfer lab ESA_Lab@DFKI, which was founded at the beginning of the year, also provides the perfect framework to explore further technology needs in ESA projects and help shape modern space applications and resulting business models with the latest AI methods.”
DFKI Project Manager Marlon Nuske explained the Solution Factory’s model: “The central idea is to leverage synergies from the development of AI solutions for various Earth observation use cases. While each individual solution will be unique in its applications, many of the underlying building blocks can be reused for additional products and projects.”
“The support from InCubed will inject some essential financial impetus into the AI4EO Solution Factory,” Mr. Nuske continued. “We’re officially kicking off this week, with several leading industrial corporations already on board for the first joint product development in the agricultural sector.”
ESA Technical Officer Nicolas Longépé was enthused by the Solution Factory’s prospects: “This is the largest ever InCubed initiative with a German partner, and we’re certain the AI4EO Solution Factory will produce important advances in AI-driven EO applications over the next three years and beyond. The genesis of the collaboration, initially between DFKI and the Φ-lab Explore Office and now with InCubed, is a salient example of our ‘innovate and apply under-one-roof’ approach.”
Timed to coincide with this year’s Φ-week event, ESA InCubed and Surrey Satellite Technology Ltd (SSTL) have announced a new initiative to design and test a high-throughput payload downlink chain for small Earth observation satellites. The module aims to provide data processing and transfer capabilities that are future-proofed for tomorrow’s onboard imagers.
Current and even more future satellite imaging payloads are generating multi-dimensional, high-volume data that can exceed the throughput capacity of small satellites. To tackle this data bottleneck, a consortium of the University of Surrey, Craft Prospect Limited and SSTL has been formed to develop product enhancements for the latter’s Flexible & Intelligent Payload Chain (FIPC) solution.
The advanced hardware architecture of SSTL’S FIPC will provide the basis for a new intelligent and adaptive data downlink, along with a state-of-the-art framework for software-defined data processing onboard the satellite. Processing tasks may include data calibration, data compression, image thumbnailing and Machine Learning (ML) for image classification.
SSTL Managing Director Phil Brownnett set out the vision for the leading-edge payload chain: “The FIPC solution will significantly improve the capabilities of small Earth observation satellites by increasing data rates to match the capacity of the onboard imager system. This will boost both the volume and complexity of the data that can be downlinked on a single satellite pass – a step up in performance that will match our customers’ business plans and ambitions.”
SSTL, Craft Prospect Limited and the University of Surrey will work together on the FIPC solution and onboard processing applications. Following the design phases, the consortium will demonstrate and evaluate the unit’s features on a flight-representative testbed.
Carlos Urbina Ortega, ESA Technical Officer for the activity, sees a number of advantages to be gained from the FIPC: “We envisage substantial benefits from this InCubed investment, not least because of the system flexibility and computational power that the FIPC affords. Capabilities such as onboard ML acceleration for enhanced data management and services, together with in-orbit application reconfiguration, will ensure that small-satellite missions remain agile and end-user focused.”
Kicking off with a bold flourish, Φ-week 2021 promises to bring space even closer to the forefront of addressing society’s biggest challenges, namely issues associated with the climate crisis, while boosting the economy through transformative New Space, artificial intelligence, and quantum and cognitive computing.
ESA’s Φ-week is now in its fourth edition, and each year this forward-looking event is increasingly relevant to society, business and the economy.
In his opening address, Josef Aschbacher, ESA’s Director General said, “Europe faces unprecedented societal, economic, and security challenges. Space has enormous untapped potential to play in tackling pressing current and future crises, while simultaneously providing new impulse for the European space sector. We need to develop smart, automated services and applications to protect lives of people and to increase the protection of assets and natural resources.”
Focusing on the New Space economy and innovations in Earth observation, ESA’s fourth Φ-week kicks off on Monday 11 October. Join us live for two of the main sessions: the Opening session on Monday at 10:30 CEST and the Blending New Space Technologies and Services session on Tuesday at 16:00 CEST.
As the world starts to return to normality in the wake of the COVID pandemic, this year’s Φ-week is a hybrid event, with registered participants attending virtually and invited speakers present at ESA’s Centre for Earth Observation in Italy.
However, part of this not-to-be-missed event can be enjoyed by all – simply tune into ESA’s Web TV and following the live transmission links.
