BridGEO

Objectives of the Product

BridGEO addresses a common challenge faced by public authorities, planners, and institutions in emerging and developing countries: critical environmental and urban decisions rely on complex satellite data that are fragmented, difficult to interpret, and often inaccessible to non-expert users. As a result, water management, urban growth control, and risk mitigation are frequently based on incomplete or disconnected information.

BridGEO provides an online, WebGIS-based platform that integrates satellite and local geodata into a unified, user-friendly system tailored to each country’s context. Its modular EO-based analytical services allow users to select affordable, fit-for-purpose solutions covering water body dynamics, imperviousness and urban sprawl, and ground motion or slope stability. A map-centric interface, cross-service analysis tools, and contextual interpretation layers transform raw data into clear, actionable insights without requiring advanced technical expertise.

To ensure long-term impact, BridGEO complements its tools with targeted, hands-on training for non-technical and technical staff alike. This builds local capacity to operate the platform independently, interpret results confidently, and apply evidence-based insights to urban planning, water management, and environmental governance, supporting sustainable and locally driven decision-making.


Customers and their Needs

BridGEO targets national and local public authorities, research institutes, and regional planning bodies in emerging and developing countries, such as those in Ethiopia, Peru, and Chile. These users are directly involved in territorial monitoring, disaster risk reduction, water resource management, and urban planning. They use the BridGEO web-based platform to routinely monitor ground movements, surface water dynamics, and land consumption, supporting early warning systems, risk mapping, and evidence-based planning decisions.

Their needs include accurate and frequent detection of landslides, the ability to analyse ground deformation trends over time, regular monitoring of reservoir volume and extent, and reliable identification of urban expansion and loss of agricultural land. They also require improved skills in GIS and satellite data analysis to integrate EO information with socio-economic and administrative data.

Meeting these needs is challenging due to limited financial resources, fragmented and inconsistent historical data, insufficient data resolution or update frequency, and a shortage of trained personnel. Traditional field surveys are costly and slow, while many EO solutions are complex and require expert knowledge. BridGEO addresses these challenges by providing affordable, integrated services and hands-on training that enable institutions to independently apply EO data in their daily operations.


Targeted customer/users countries

Targeted countries are Ethiopia, Peru, and Chile.


Product description

BridGEO is an online, WebGIS-based platform that delivers integrated Earth Observation services for territorial monitoring, risk management, and environmental governance. The system combines three core analytical services within a unified environment. The Water Bodies Dynamics service provides monthly monitoring of reservoirs in user-defined areas, delivering trend analysis and change detection at 10 m spatial resolution and a minimum mapping unit of 0.5 ha. The Imperviousness Dynamics service classifies and monitors land cover using an adapted EAGLE Level 2 scheme, generating six-monthly updates and automated land-change maps at the same spatial resolution. The Ground Motion service delivers trends in ground elevation change over a three-year window, updated every six months, with outputs available at native resolution and resampled grids of 50 m or greater.

The main innovation lies in the integration of these services into a single, map-centric web interface. Users interact through intuitive dashboards, predefined thematic views, and automated cross-service analyses that reveal correlations between urban growth, water dynamics, and ground instability. A contextual interpretation layer translates technical results into plain-language insights. BridGEO is complemented by modular, hands-on training—delivered on-site or remotely via an AI-supported LMS—enabling users to independently analyse data and apply results in real-world decision-making.


Added Value

BridGEO delivers significant added value compared to existing EO-based monitoring methods by combining higher-resolution data, more frequent updates, and integrated analysis within a single operational platform. For water bodies monitoring, BridGEO goes beyond commonly used approaches—such as reliance on a single index like NDWI—by generating inland surface water detection maps based on multiple multispectral indices. This improves detection accuracy and robustness across different environmental and seasonal conditions. The service also provides water occurrence and frequency maps, distinguishing maximum, seasonal, and permanent water bodies, as well as inter-annual and seasonal variation maps that support informed management of agricultural activities during dry and wet periods.

Compared to existing national and international services, BridGEO offers capabilities that are currently unavailable or limited in many emerging countries. High-resolution imperviousness maps enable detailed urban analysis and detection of subtle land cover changes, addressing a major gap in countries such as Ethiopia. Relative to Copernicus services, BridGEO delivers higher spatial resolution, more frequent updates, and a tighter integration between imperviousness analysis and next-generation IRIDE data. By integrating these advanced datasets into a unified, user-friendly system, BridGEO provides more actionable, locally relevant insights than standalone or medium-resolution EO services.


Current Status

The project has made significant progress in assessing current conditions, identifying vulnerabilities, and spotting opportunities, establishing a solid foundation for prioritising services. Requirements definition is underway, with user needs, system specifications, and service expectations being gathered, analysed, and documented to ensure clarity of functional and non-functional requirements.

Technical development has commenced, focusing on the design, construction, and testing of EO services, the delivery platform, and associated training courses. Work in progress includes refining technical solutions, validating system functionalities, and preparing training materials. Upcoming activities will concentrate on finalising requirements, completing system testing, and rolling out training programmes. Overall, the project is advancing steadily, with foundational analysis and technical development progressing in parallel to ensure services are well-aligned with user needs and operational priorities.

EARS-GNSS RO

Objectives of the Product

GNSS Radio Occultation (GNSS RO) is a space-based remote sensing technique for studying the Earth’s atmosphere. It uses the atmosphere’s refractive properties, which bend radio signals from GNSS satellites as they reach a low Earth Orbit (LEO) receiver. This bending causes an additional Doppler shift, which, when compared to the expected shift in vacuum, allows the atmospheric contribution to be used to derive the bending angle.

The bending angle (along with impact parameter) is a key radio occultation observable, forming the basis for retrieving vertical profiles of atmospheric refractivity, temperature, pressure, and humidity. GNSS RO provides high vertical resolution, global coverage, and long-term stability, making it valuable for numerical weather prediction (NWP). Minimising latency between measurement and data assimilation is crucial for applications like nowcasting.

The Earth Atmospheric Remote Sensing via GNSS RO (EARS-GNSS RO) mission is developing a LEO satellite capable of determining key parameters of Earth’s atmosphere with high vertical resolution and low latency using the GNSS RO technique. Key aspects of the mission are:

  1. On-board processing: The mission uses a high-performance System-on-Chip (SoC) and an advanced Precise Orbit Determination (POD) algorithm, enabling on-board data processing.
  2. Latency Reduction: Thanks to onboard data processing and a network of low data-rate ground stations, the latency of transmitted information is reduced.

Customers and their Needs

The primary users of GNSS RO data from the EARS mission include:

  1. Numerical Weather Prediction (NWP) centres
  2. Meteorological agencies
  3. Climate monitoring institutions
  4. Commercial organisations relying on weather and atmospheric intelligence

These users engage with the activity through data assimilation, quality control feedback, and validation campaigns, integrating the data into forecasting, nowcasting, and climate monitoring workflows.

Users need timely, accurate atmospheric profiles with precise geolocation and low latency, along with reliable coverage and products compatible with existing operational and commercial systems.

Challenges include maintaining low latency and high quality, as well as managing the extra power required for on-board processing. In addition, the implementation of on-board precise orbit determination (POD) to compute real-time satellite positions with sufficient accuracy to ensure the quality of the calculated bending angles and atmospheric profiles poses further challenges.


Targeted customer/users countries

The targeted customers/users are both commercial and scientific, spanning Europe, North America, and Asia.


Product description

The primary EARS product is low-latency, high-quality atmospheric data. The system is composed of the following main building blocks:

  1. EARS Platform, hosting the RO instrument and a secondary payload, providing the necessary spacecraft resources and interfaces.
  2. RO Instrument, including a GNSS receiver, POD module, and dedicated antennas for GNSS RO measurements.
  3. Ground Segment, responsible for telemetry, tracking and command (TT&C) operations and low-latency data downlink and distribution.
  4. Cloud Services, enabling scalable and user-friendly access to RO data products for operational and commercial users.

A block diagram of the complete end-to-end system, including the GNSS RO measurement context described in the product objectives, is shown below:

The EARS system combines advanced algorithms with optimised on-board hardware to enable Near Real-Time (NRT) processing of measured data directly on orbit. This innovative approach optimises downlink usage and significantly reduces data latency, allowing faster delivery of user-ready atmospheric products. Users interact with the product through cloud-based services, integrating the data directly into forecasting, nowcasting, and atmospheric analysis workflows.


Added Value

The proposed solution delivers added value compared to existing GNSS RO services through innovations in processing, platform architecture, and European commercial sourcing.

Onboard near-real-time processing sets this solution apart from current missions relying on ground-based processing. By performing signal acquisition, processing, and atmospheric profile derivation on board a small satellite, data latency is reduced from hours to under 30 minutes. This enables operational use cases such as nowcasting, rapid-response meteorology, and aviation weather services.

Compact, scalable platform architecture based on a high-performance Software Defined Radio (SDR) with integrated on-board POD and processing allows deployment on small platforms, e.g., 8U-12U CubeSats. Distributed constellations can provide higher spatial and temporal coverage at lower cost.

European-origin commercial data source – the system enhances strategic autonomy, offering a resilient alternative to non-European providers and supporting institutional, regional, and potential dual-use applications.


Current Status

The EARS project is currently in the de-risking phase, focusing on a detailed analysis of the market and customer requirements and the mission, payloads and system-level requirements. The subsequent activities focus on the definition of preliminary requirements and the architecture for the mission, payload, platform, and ground segment.

AI Visual Copilot

Objectives of the Product

Organisations across sectors such as agriculture, forestry, infrastructure, and insurance face major barriers in applying Earth observation (EO) data. Creating tailored AI models for tasks like object detection or segmentation requires costly manual labelling, deep technical expertise, and months of development time. As a result, many EO projects are delayed, scaled back, or abandoned.

FlyPix AI addresses these challenges with the AI Visual Copilot, a no-code SaaS solution that streamlines EO model creation. By combining one-shot learning with AI-assisted annotation, the Visual Copilot reduces labelling efforts by up to 95% and cuts model development time from months to days. Users without AI expertise can interactively define models, leverage smart labelling suggestions, and deploy fit-for-purpose solutions at scale.

The Visual Copilot integrates seamlessly into the FlyPix AI platform, supporting diverse EO data sources from satellites, drones, and aerial imagery. Its modular design ensures scalability, adaptability, and rapid prototyping for a wide range of applications, from land use monitoring and biodiversity protection to damage assessment and infrastructure tracking.

This solution empowers customers to transform raw EO data into actionable insights faster, cheaper, and with greater flexibility, unlocking new opportunities for data-driven decision-making across industries and public authorities.


Customers and their Needs

The targeted customers include geospatial and geographic information systems (GIS) service providers, aerial inspection firms, insurers, public authorities, and environmental agencies. These organisations rely on EO data to monitor assets, assess risks, track environmental change, and support critical decisions. However, they face significant challenges in leveraging EO data effectively.

Key needs include:

  • Faster model creation: Current workflows require extensive manual labelling and weeks of preparation, delaying projects.
  • Lower costs: Tailored AI models demand high budgets, often consuming most of a project’s resources.
  • Ease of use: Many customers lack in-house AI expertise, making advanced EO analytics inaccessible.
  • Scalability: Models are difficult to adapt across datasets, regions, or new use cases.
  • Confidence in ROI: High upfront investment and uncertain outcomes discourage adoption, especially among SMEs and public bodies.

Through the activity, pilot customers such as insurers, GIS firms, and aerial inspection providers will directly test the Visual Copilot, providing feedback on workflows and usability. By enabling no-code, AI-assisted model creation that reduces costs and time by up to 95%


Targeted customer/users countries

Based on the proposal, the targeted customer/users’ countries are Germany, Austria, Netherlands, and the rest of EU countries; United Kingdom, USA, Japan and India.

In addition, the product is positioned for broader uptake across Europe, with future expansion to North America and Asia-Pacific.


Product description

FlyPix AI Visual Copilot is a no-code, cloud-based solution that accelerates the creation of AI models for EO data. It enables users to perform object detection, segmentation, and tracking on satellite, aerial, and drone imagery with minimal technical expertise. Customers interact through an intuitive interface where they can define tasks visually, apply AI-assisted labelling, and deploy models within hours rather than months.

Innovation lies in the one-shot learning approach, enabling model creation from a single example, combined with automated labelling and modular scalability. This transforms EO analytics into a fast, affordable, and user-friendly process, unlocking new opportunities for industries such as agriculture, forestry, insurance, and infrastructure monitoring.


Added Value

FlyPix AI Visual Copilot delivers a step-change in how organisations build AI models for Earth observation by eliminating the biggest barriers—cost, time, and lack of expertise. Competing platforms require large, annotated datasets, weeks of preparation, and advanced technical knowledge, making EO-driven insights inaccessible to many organisations.

In contrast, FlyPix AI Visual Copilot allows users to build accurate detection or segmentation models from just a single example. Combined with AI-assisted annotation, it reduces labelling efforts by up to 95% and lowers project costs by as much as 70%. This makes EO analytics affordable for small enterprises, local authorities and NGOs, in addition to large commercial players.

Unlike competitors focused on narrow use cases, Visual Copilot offers broad versatility—supporting tasks from agriculture monitoring and forestry management to insurance damage assessment and infrastructure tracking. Its no-code interface empowers non-experts to create, validate, and deploy models quickly, accelerating decision-making in critical areas such as climate resilience, urban planning, and disaster response. The added value lies in democratising EO model creation: transforming months of expert-driven work into a simple, scalable, and cost-effective workflow that unlocks new opportunities for innovation and impact across industries.


Current Status

AI Visual Copilot activity builds on a working prototype already integrated into the FlyPix AI platform. Initial proof-of-concept projects have been successfully delivered for customers in insurance, forestry, and infrastructure, validating both the technical approach and market demand. Letters of Interest (LoIs) from pilot customers confirm strong commitment to adopt the solution once available.

Currently, the team is refining the prototype into a scalable SaaS module, focusing on one-shot learning, AI-assisted labelling, and seamless EO data integration. Pilot users are engaged in requirements gathering and early validation to ensure usability and alignment with operational needs. The upcoming phase will expand testing with multiple pilot customers, finalise integration with EO data providers, and prepare the platform for commercial rollout. With InCubed’s support, the activity is on track to deliver a validated, market-ready product by early 2026.

ENGINE

Objectives of the Product

The main problem in the Earth observation (EO) market is that the technology used is too complex and requires specialised knowledge for most users to interpret satellite data and integrate it into their operations, hindering wider adoption.

The solution is the ENGINE project, an AI-powered agent integrated into the SpaceKnow Guardian® platform. Its objective is to democratise geospatial intelligence, initially focusing on urban development and construction monitoring.

ENGINE achieves this by:

  • Allowing users to control the analysis platform and receive results using simple natural language.
  • Eliminating the need for remote sensing expertise.
  • Combining imagery intelligence with contextual data from open-source intelligence (OSINT) to deliver actionable, all-in-one insights.

This integration streamlines the entire analysis process, making powerful, scalable EO insights accessible to a much broader range of businesses and organisations.


Customers and their Needs

The targeted primary customers for the activity are from the urban planning and construction monitoring segments. Other key segments include national security entities, NGOs, and the oil and gas industry. The long-term vision is to expand to virtually any organisation that can benefit from scalable satellite imagery analysis.

Customers face challenges due to the complexity of integrating Earth observation (EO) solutions and the need for specialised, technical expertise to interpret geospatial data. This creates a steep learning curve and limits adoption. Users interact with the product, ENGINE, an LLM-powered autonomous agent, by using natural language (written or spoken) to control the system and request analyses. This involvement allows them to autonomously monitor projects and receive actionable insights and contextual information without needing expertise in remote sensing.


Targeted customer/users countries

Europe, US, KSA.


Product description

The product is ENGINE (SpaceKnow Guardian® Operator), an LLM-powered autonomous agent integrated into SpaceKnow’s cloud-based platform, SpaceKnow Guardian®.

Capabilities and Innovation:

ENGINE’s core capability is democratising geospatial intelligence. It transforms the analysis of satellite images (IMINT) by combining it with contextual data from open-source intelligence (OSINT). The innovation lies in the LLM-powered agent, which simplifies the entire geospatial analysis workflow, dramatically reducing the technical knowledge required from the user. It also employs proprietary algorithms to process challenging low-resolution SAR images.

User Interaction and System Outline:

The user interacts with the product using natural language (written or spoken inputs). The system processes the request and delivers results as automated responses in an LLM-processed format (written, spoken, or visual).


Added Value

The proposed solution, ENGINE, brings added value and a competitive edge through several key innovations:

  • LLM-Powered User Assistance: The main innovation is an LLM-powered agent that assists the user, reducing the entry barrier for geospatial analysis. It suggests the most suitable approach and simplifies the workflow, a significant advancement over current offerings.
  • Data Fusion: It achieves optimal, customised results by merging Earth observation (EO) data with other sources like open-source intelligence (OSINT), a capability uncommon in the industry.
  • Superior Processing: Proprietary algorithms are even able to process challenging low-resolution Synthetic Aperture Radar (SAR) images, addressing data quality difficulties that may limit competitors.
  • Streamlined Service: By employing AI agents for both user input and output interpretation to add contextual knowledge, ENGINE is not just an improved interface but an enhanced overall service, making it highly competitive against rivals like Preligens or Orbital Insights.

Current Status

The activity kicked off it the beginning of Q4 2025. Currently, the first work packages are running, focused on State-of-the-Art Review, Foundation Model and Agency Framework exploration, and External Data Research. These work packages will be reviewed in the first milestone. The Requirements’ Review is planned for 11 December 2025.

VISIONS

Objectives of the Product

The market’s insatiable demand for more data, at higher spatial and temporal resolutions, is driven by the need to gain deeper insight into the Earth’s environment and respond to the observed situations in a timely manner. In response to this, the product developed under VISIONS offers substantial performance improvements over existing products in the SSTL portfolio, providing higher resolution data, as well as significantly greater data return, and is capable of rapid tasking and product delivery. The problem to be addressed is providing the Earth Observation (EO) market with cost-effective sub-0.5 m resolution imagery, provided in a way that is timely, trusted, actionable and at a reasonable cost. The multispectral payload developed in a previous activity enables customers to benefit from actionable data and insights at very high resolution to enhance their services and business operations in a wide range of vertical markets including agriculture, disaster and resource management, defence and security.


Customers and their Needs

For Commercial EO companies, customers need cost-optimised, high-performance imaging solutions in Low Earth Orbit (LEO) to maximise profitability of their business, shorten lead times to accelerate their time to market and revenue generation, and then to maximise the utilisation of their mission in a variety of markets and applications. Commercial customers also need an ability to scale operations rapidly in response to greater market demand, driving developments toward repeat builds, less bespoke solutions, and modularity in small satellite space systems. 

Government customers (Civil and Defence) also express these commercial needs but with added pain points related to sovereignty and EO data independence, filling critical EO data gaps, capacity building and knowledge exchange, highly secure architectures with strict security and data traceability requirements, and finally value-for-money and ensuring the best use of public spending to benefit society and the economy as a whole.


Targeted customer/users countries

As a small satellite and payload manufacturing company, SSTL customers target satellite operators, across Commercial EO companies and Governments. Key target market areas include UK, Europe, Middle East and South-East Asia (EMESEA) .


Product description

The VISIONS activity directly supports the delivery of SSTL’s Precision product; a 450kg, mini class satellite, capable of being compatible with affordable rideshare launchers, and capturing very-high resolution imagery from a 500-km orbit. The Precision imager capitalises on half-pixel shifted pixels and processing in the PAN band to provide superior resolution and image quality (particularly benefiting from reduced aliasing). The half-pixel shifted pixels in the PAN band improve the spatial sampling from the native PAN GSD of 0.6m to 0.3m. In addition, the Imager can also simultaneously capture four 1.2m-GSD multispectral channels from a choice of six available (Coastal blue, B, G, R, Red Edge & NIR). The product benefits from an upgraded downlink chain in X-Band capable of an Over the Air (OTA) data rate of 900Mbps per transmitter. The modular payload and downlink chain that sits on the spacecraft platform comprises upgraded avionics modules that are designed for constellation production to meet market demands. 

The product is designed for a wide range of LEO spacecraft EO operators who can seamlessly integrate into their ground segment’s infrastructures and data service to their users and downstream value chain.


Added Value

Customers will gain combined added value over other small EO satellites on the market from the Precision product including:

  • Increased information and intelligence value per square kilometre, through ‘best in class’ High EO Imaging Performance
  • Reduced operational costs through increased downlink rate
  • Reduced costs through compact design and ability to access affordable rideshare launch options
  • Enhanced customer use and business cases through agility, tasking and data latency
  • Reduced non-revenue generating data through access to on-board processing capabilities
  • Rapid delivery, modular approach, and ready for production for constellations.

Current Status

The activity had its kick-off meeting on 23 October 2025.

Emissions Watch

Objectives of the Product

The UK government has implemented several policies and strategies to reduce methane emissions and achieve net-zero by 2050. However, without good data underpinning decision-making and action, the UK will fail to meet these targets. This solution will bridge the gap between existing satellite-based measurement data and operationally relevant, actionable information.

Greenhouse gas emissions linked to UK consumption are rising, driven largely by imported goods. Due to its high global warming potential, methane is a priority for action, as the potent greenhouse gas contributed around 16% of the UK’s territorial emissions in 2023. Agriculture, waste, and energy sectors accounted for 49%, 30%, and 8% of these emissions respectively. Methane also makes up an estimated 15% of emissions embedded in imports, largely from oil and natural gas—53 billion cubic meters imported in 2022 alone.

Over half of methane emissions are “fugitive,” meaning their sources and timing are unpredictable and so difficult to estimate or detect without frequent monitoring. Without robust data on methane sources, the UK risks missing its net-zero targets. Satellites offer a cost-effective, consistent way to for regular monitoring of methane emissions.

Emissions Watch delivers detailed mapping and monitoring of industrial greenhouse gas emissions using advanced satellite analytics and contextual facility data. It provides timely, actionable insights to support compliance, inform policy, and drive sustainable outcomes.


Customers and their Needs

Government agencies are the target market for this service—both the United Kingdom and wider Europe.

In the United Kingdom, there are four specific needs:

  • Replacing modelled estimates of methane emissions in the National Atmospheric Emissions Inventory (NAEI) with empirical, measured data
  • Tracking emissions from imports against international commitments, assessing compliance and informing future import standards
  • Supporting the implementation of methane reduction schemes and incentives, such as the expansion of the UK emissions trading scheme and carbon border adjustment mechanism, through independent Monitoring, Reporting, and Verification
  • Enforcing existing and future environmental and health and safety regulations, such as Leak Detection and Repair programs or emissions threshold penalties

Targeted customer/users countries

The service will be primarily targeted towards the UK before being out to Europe and broader global markets.

Regulatory agencies include the UK Environment Agency, Scottish Environmental Protection Agency, Natural Resources Wales, North Sea Transition Authority, and the Office of Gas and Electricity Markets.

Policy bodies include Defra and DESNZ, as well as similar agencies across European governments.


Product description

Emissions Watch will provide a comprehensive view of methane emissions tied to specific facilities, allowing the tracking of emissions at the site level—Asset-Level Analytics—and portfolio level by type, sector, ownership, or geography for Executive Reporting.

The service will initially focus on methane from onshore oil and gas and waste facilities using artificial intelligence (AI) and satellite data to identify and map facilities.

By combining information from a range of public and private data sources, it will then connect these mapped locations with important details such as the type of facility and ownership information, creating a “Facility database”.

Finally, Emissions Watch will draw on remotely sensed satellite short-wave infrared data (GHGSat and Sentinel-2) to both detect and measure emissions and then link those measurements directly to the corresponding industrial facilities via machine learning.


Added Value

The key areas of innovation brought by Emissions Watch include:

  • Greater accountability for emissions variability and unexpected sources. Augmenting existing estimates of methane emissions modelled by customers with observed empirical data (from satellite) will improve the overall accuracy of emissions recorded. This is done by addressing inherent issues of error and uncertainty caused by the dependency on generalised emissions factors and activity data used in modelling. Modelling approaches fail to capture super-emitter events or operational anomalies that contribute disproportionately to total emissions (up to 60% according to different sources). Furthermore, independent mapping of methane emitting facilities (using satellite and AI) will ensure full transparency and accounting for all emitting facilities, minimising non detection or underestimation that results from fugitive emissions which are typically an important part of the total emissions profile.
  • Linking emissions activity to the accountable party. Providing origin certainty and linking probability of ownership at the source of emissions will allow relating of observation data through to relevant regulatory or licensing frameworks to support decision making and action.
  • Comprehensive global reference for observed emissions data. A complete emissions dataset (enabled by combining GHGSat with public sector data) uniquely supports analytics and reporting that allows an individual facility to be put into the context of global trends (By facility type, sector, ownership or geography). This allows for performance measurement to support applications such as tracking reduction targets or benchmarking against competitors.

Current Status

The activity held its Kick-Off meeting on the 4 September 2025.

DVSTAI

Objectives of the Product

Developing and deploying neural network-based computer vision models for geospatial data is challenging for most organisations. Typically, AI experts customise a model architecture and train it using specific datasets provided by the customer, aiming to detect predefined targets. After the model is trained, it is deployed as a “black box,” limiting the user’s ability to improve detections or add new targets without extensive re-engineering and re-training, which incurs significant time and costs. This process is not only inefficient but also poses security risks, as training data must be accessed by external engineering teams, complicating data traceability and increasing the risk of unauthorised data access and leakage. Furthermore, the lack of thorough traceability can lead to data poisoning, facilitating various attacks like adversarial attacks, label flipping, and model inversion, which can undermine the model’s integrity and potentially expose original training data.

Building on the precursor de-risk activity “SatHound: Multi-object detection solution based on artificial intelligence for non-expert EO users – eo science for society”, DVSTAI offers a new Earth Observation (EO) tool to ease the daily work of EO analysts and non AI experts, an object detection solution that end-users can configure autonomously by creating, training and deploying their own AI models to perform vision tasks over satellite imagery.


Customers and their Needs

Target customers are:

  • Governmental agencies
  • EO Satellite operators
  • EO Service providers
  • EO Analysts

DVSTAI targets niche markets of governmental and private entities operating Earth Observation satellites that require user-level analysis tools to obtain fast results on specific tasks. It is expected that such customers are owners of geospatial data and expect to use DVSTAI within their own infrastructure, potentially integrated with other processing tools in their data management process.

DVSTAI also targets to address mass markets via EO service providers, that can offer both data provisioning and analysis tools. DVSTAI can be offered bundled with existing EO services to enable their customers to create specific vision tasks over available data sets in the platform to extract additional value while keeping full flexibility of a Software as a Service model.


Targeted customer/users countries

DVSTAI targets the customer segments described above globally.


Product description

The product is designed to enable end-users to perform object detection on various types of satellite images across different spatial and temporal selections from the catalogue. It features a graphical interface that allows users to label objects of interest, train new or refine existing specialised detection models, manage trained models, administer catalogues, and conduct searches within the catalogue using one or more specific models.

The main differentiation objectives are of DVSTAI are:

  • Allowing end-users to create and manage object-detection and other vision neural networks trained to detect their target of interest in satellite imagery
  • Allowing end-users to perform target detection for multiple objects in the same area of interest and on multiple search areas simultaneously
  • Allow end-users to monitor an area of interest over time, by scheduling the processing of new images available in the catalogue and easily comparing results with previous searches.
  • The tool is usable by non-expects (in Artificial Intelligence, ML pipelines or remote sensing) with a good user experience
  • The user can work with public data or their own optical, multispectral, SAR; and manage different band combinations
  • The tool supports different AI model architectures and is extendable for new ones
  • Allow Software as a Service pay-per-use model by data segregation, customer on-boarding and multi-tenancy

Added Value

The idea behind DVSTAI is to ease the daily work of EO analysts, but also to offer the general public an object detection solution that end-users can quickly set-up and use as a service to analyse public geospatial or their own data.

The solution offers an end-to-end architecture for data management, labelling, training and inference, that allows users with no or limited knowledge of AI and EO business to implement their own models on top of the platform. DVSTAI models can be tailored to fit any specific use cases where vision is involved such as object detection, change detection, semantic segmentation, etc.

DVSTAI provides added value to its customers by:

  • Reducing the knowledge gap with a tool where domain knowledge is the only thing needed to create/configure AI vision tasks.
  • Facilitating trainings for new targets after annotating a few tens of representative samples. Re-trained models are automatically ready for production use.
  • Providing a unified set of performance metrics, automatically generated on each new version of the model after training. Its model version management allows to roll-back to previous versions if required.
  • Managing data governance by secure access control to the solution HMI and APIs. Removing the need for manual data extraction and processing outside the platform that could lead to leakage.
  • Native integration with service providers in cloud-based platforms that offer collections of data catalogues from open and commercial sources

Current Status

The DVSTAI InCubed project started in June 2025 and the first project milestone (MVP definition) is scheduled for September 2025.

HAPSEYE

Objectives of the Product

HAPSEYE is an ESA InCubed-supported project led by ICEYE Spain to develop a solar-powered High Altitude Pseudo-Satellite equipped with a SAR payload. Operating at more than 20 km in altitude, HAPSEYE complements ICEYE’s satellite constellation by providing continuous monitoring with near real-time data transmission.

Through a series of incremental prototypes, the project advances towards a Minimum Viable Product (MVP) at TRL7. HAPSEYE enhances Europe’s autonomy in Earth observation, bridging gaps left by satellites due to limited revisit rates or capacity bottlenecks. Its applications include disaster response, flood and wildfire monitoring, and national security.

This activity consolidates ICEYE’s Spanish hub in Valencia, creating new high-skilled jobs and strengthening the European aerospace supply chain. With support from ESA, national regulators and key customers HAPSEYE is positioned to become the first European SAR-tailored HAPS platform, delivering unique added value in resilience, flexibility, and cost-effectiveness.


Customers and their Needs

HAPSEYE targets government agencies, emergency management organisations, insurers, reinsurers, and defence entities. These customers require persistent, timely, and reliable Earth observation data to support disaster management, infrastructure monitoring, and security operations.

Insurance and re-insurance customers have expressed strong interest in validating HAPSEYE results during development. Governments require enhanced situational awareness during crises.


Targeted customer/users countries

Spain, Greece, USA, Australia, Japan.


Product description

HAPSEYE is a solar-powered unmanned aerial platform designed to operate in the stratosphere for months at a time. It carries a Synthetic Aperture Radar payload, providing near-continuous coverage of selected regions with very high resolution.

Main features:

  • Continuous coverage above 20 km altitude, staying out of adverse weather conditions.
  • Flexible deployment, loitering above crisis areas, and integration with ICEYE’s satellite fleet.
  • SAR data collection and delivery within ~3 hours when combined with ICEYE ground segment.
  • MVP platform capable of year-round flight and extended operational latitudes

Innovation aspects: dedicated SAR-only payload, agile prototyping cycles, and European development (avoiding ITAR restrictions). Customers interact via ICEYE APIs, portals, or tasking services, integrated with existing data pipelines


Added Value

HAPSEYE uniquely combines persistence, high-resolution imaging, and resilience:

  • Versus satellites: Provides continuous monitoring of a region instead of periodic revisits.
  • Versus aircraft/UAVs: Much longer endurance (months), lower operational costs, and weather resilience.
  • Versus competitors: SAR-only focus ensures optimisation for radar missions

The integration with ICEYE’s SAR constellation and Solutions products makes HAPSEYE a key enabler of faster disaster response and improved risk assessment. Its European development secures autonomy in critical Earth observation capacity.


Current Status

Kick-off meeting held on 30 July 2025.

Clear Sky Constellation

Objectives of the Product

A key unmet demand from Oil & Gas companies is the reliable and easy detection of methane emissions so:

  • Companies can comply with worldwide regulations
  • Companies can take responsibility

Airbus aims to offer reliable and easy worldwide monitoring of methane emissions (onshore and offshore) using a fleet of satellites.


Customers and their Needs

The primary customers and users that are targeted are Oil & Gas companies, governmental organisations (the European Space Agency/European Commission, national governments), finance sector and scientists.

The issue is that customers are not satisfied with the existing satellite services for greenhouse gas (GHG) emissions monitoring, which do not reach the detection limits needed to detect the actual leaks in their facilities.

The primary customers participate in a series of collaborative workshops to help shape a service that will enhance their operational efficiency and environmental stewardship, while reducing the overall costs of emission detection.

The main challenge is to provide high resolution methane emission monitoring at affordable cost.


Targeted customer/users countries

The service is offered to customers all around the world.


Product description

Airbus aims to offer reliable and easy worldwide monitoring of methane emissions (onshore and offshore) using a constellation of small satellites.

The innovation aspect that Airbus is focusing on is the space-based monitoring at unprecedented resolution and at viable cost.

How it works:

  • A customer subscribes to the monitoring of a facility(-ies)
  • Airbus provides regular methane emission measurements for that facility(-ies) in line with the service level set (sensitivity, false detection rate, localization, frequency).

The product/system architecture is:


Added Value

Key Benefits:

  • Global routine monitoring, fully automated
  • Reliable space-based detection at unprecedented resolution
  • Effortless access, everywhere (e.g. remote, inaccessible, non-operated sites)
  • Adapt with ease to facility park changes
  • Consistent data for tracking long-term progress
  • Facility-wide benchmarking: consistent, comparable, clear
  • Preconfigured reporting: speed with precision
  • Enhancing workplace safety with automation
  • Methane and CO2, detected together.

Current Status

By teaming up with SRON, Airbus is developing a constellation of small satellites that can measure methane emissions at facility level with unprecedented resolution. Both are world leaders in the domain of space based atmospheric monitoring with a proven track record in methane emissions monitoring (e.g. ESA Sentinel-5p TROPOMI)..

LTA

Objectives of the Product

Property tax in Poland and other SCBE countries (Slovakia, Czechia, Bulgaria, Estonia) relies on self-declarations, which often leads to underreported property size or unregistered buildings. As a result, municipalities lose significant revenues, limiting their ability to invest in public services and infrastructure.

LandTaxAssessor addresses this problem by providing local governments with a reliable way to verify reported land use. The solution combines satellite Earth Observation (EO) data with geodetic maps and applies advanced AI algorithms to identify discrepancies, such as undeclared buildings or underestimated property areas.

The objective of the product is to reduce tax evasion, increase municipal revenues, and ensure fair taxation across communities. By delivering fast, regulation-compliant reports through an easy-to-use online portal, municipalities can act quickly without needing in-house expertise in geospatial technologies.

Ultimately, LandTaxAssessor lets local administrations modernise their tax collection processes, strengthen transparency, and recover substantial funds.


Customers and their Needs

The primary customers of LandTaxAssessor are local governments and municipal tax departments in Poland and the SCBE countries (Slovakia, Czechia, Bulgaria, Estonia).

These institutions are responsible for calculating and collecting property taxes, yet they face significant challenges when relying on self-declarations from property owners. Undeclared or underestimated property size, as well as unregistered buildings, create a gap between real land use and reported tax obligations.

The key customer need is a reliable, cost-effective, and user-friendly method to detect tax evasion without requiring in-depth expertise in Earth Observation or GIS. Municipalities need a solution that fits seamlessly into their existing administrative workflows, provides legally compliant results, and delivers them within a short timeframe to support timely decision-making.

With LandTaxAssessor, customers gain access to an online portal where they can request property tax assessments. The service returns AI-driven analyses and regulatory reports, helping local governments identify discrepancies and recover missing revenues. This enables municipalities to increase their budgets and to ensure fairness and transparency in taxation.


Targeted customer/users countries

The targeted customers are local governments and municipal tax departments in Poland and the SCBE region, specifically Slovakia, Czechia, Bulgaria, and Estonia. These countries share a similar property tax system based on self-declarations from property owners, which makes them equally exposed to risks of underreporting and unregistered buildings. The solution is designed to be directly applicable across these national contexts.


Product description

LandTaxAssessor is an online service designed for local governments to improve property tax collection. It integrates Earth Observation (EO) satellite imagery with official geodetic maps and applies advanced AI algorithms to detect discrepancies such as unregistered buildings, underestimated property areas, or changes in land use.

The system is accessed via a secure web portal, optimised for administrative users with no background in EO or GIS. Municipalities can simply submit an analysis request, and within less than four weeks they receive a regulation-compliant report that highlights potential cases of tax evasion. Reports are clear, structured, and ready to support administrative procedures.

A key innovation of LandTaxAssessor is the automated fusion of EO data with cadastral maps and the use of AI-driven building detection. This enables fast, scalable, and accurate assessments that would otherwise require significant time and specialised staff.

From a high-level architecture perspective, the system consists of:

  • A data ingestion layer (EO imagery and geodetic maps).
  • AI-based processing for building extraction and land use classification.
  • A cloud-based portal (NSIS marketplace) for service requests and delivery.

This approach provides a modern, efficient, and affordable tool for municipalities to strengthen transparency and fairness in tax collection.


Added Value

LandTaxAssessor brings significant added value compared to traditional methods and existing tools. Today, property tax verification often relies on manual inspections, fragmented cadastral data, or specialised GIS teams, which are costly, time-consuming, and difficult to scale. As a result, many municipalities cannot effectively address underreporting or unregistered buildings.

LandTaxAssessor overcomes these limitations by combining EO satellite imagery with geodetic maps in an automated, AI-driven process. This unique fusion enables fast and accurate detection of discrepancies without requiring in-house GIS or EO expertise. Municipalities access the service through a simple online portal, making the process transparent, user-friendly, and affordable.

Another advantage is turnaround time: LandTaxAssessor delivers complete, regulation-compliant reports in less than four weeks. Competing methods often require lengthy fieldwork or multiple data providers, which delays action. By contrast, LandTaxAssessor’s cloud-based approach scales easily across regions and municipalities, ensuring consistent quality at lower cost. Finally, the solution directly supports sustainable urban development.

By helping cities recover millions of euros in lost tax revenue, LandTaxAssessor lets governments reinvest in infrastructure and community services. The added value comprises improved efficiency and strengthened fairness, transparency, and public trust in taxation.


Current Status

After the Project Kick-Off on 31 July 2025, the partners launched coordination activities and held several technical meetings to align tasks and responsibilities. Consultations with three pilot towns (Kwidzyn, Piaseczno, and Wołomin) were carried out to better understand their daily operations and expectations. These discussions helped define the final set of use cases.

To gather detailed input, questionnaires and mini-workshops were organised. As a result, two baseline scenarios were developed, focusing on detecting changes in buildings and in land usage registers. The collected requirements, including legal and technical aspects, were analysed and turned into system specifications and design guidelines.

An initial testing programme has also been outlined, and a review of available methods, datasets, and legal regulations has been completed. The first formal review meeting is scheduled for 29 September 2025. The team is now preparing data and starting development of the core algorithms and system modules.