Unlocking the Potential of Spatial Data

The Dynamic Vicmap Project

In today’s rapidly changing technical landscape, a major challenge exists in how to use space-based data streams to update critical digital infrastructure rapidly, efficiently, and cost-effectively, while maintaining the quality and trust in that data.

Currently, the potential of space-based data remains largely untapped in Victoria. The Dynamic Vicmap project will leverage new artificial intelligence (AI) capabilities and space-based data streams to showcase how foundational spatial data can be managed to provide rapid automated updates that are dynamic and comprehensive. Data is enriched with contextual information about what is changing and why, ensuring that data changes are traceable, allowing deeper insights and more informed decision making.

The Dynamic Vicmap Project is a groundbreaking initiative co-funded by a grant from SmartSat VIC Node: Space Demonstrator Program and supported by the Victorian Government Department of Transport and Planning. Led by FrontierSI, it is a collaborative effort with the Victorian Government Department of Transport and Planning, and RMIT University.

The Test Case: Vicmap Hydro

Using the Vicmap Hydrology theme as a test case, the project explores opportunities for innovation through machine learning and new data models. Dynamic Vicmap leverages AI techniques, knowledge graphs, automatically generated linked temporal data, and international standards to enrich data with context (also known as semantic enrichment).

Semantic data integration is the process of combining data from disparate sources and consolidating it into meaningful and valuable information through the use of Semantic Technology.

Semantic data integration is the process of combining data from disparate sources and consolidating it into meaningful and valuable information through the use of Semantic Technology.

For example, machine learning is used to capture the current waterline of a reservoir from satellite imagery. Automatically generated linked data provides background information such as the date and source of imagery, as well as the machine learning training accuracy. This contextualises that update for users and supports potential updates to existing authoritative representations of the boundary of a waterbody.

Water reservoirs represented with different geometries: (i) yellow polygons representing authoritative Vicmap water features, (ii) blue polygons representing para-authoritative water features generated by a ML model, (iii) green points representing authoritative Vicmap water features.

Water reservoirs represented with different geometries: (i) yellow polygons representing authoritative Vicmap water features, (ii) blue polygons representing para-authoritative water features generated by a ML model, (iii) green points representing authoritative Vicmap water features.

Far-reaching Impact

The success of the Vicmap Hydrology test case could be the catalyst for broader change in the management of foundational spatial data. It signifies the potential of enriched data, enabling the same transformative process for a myriad of other datasets. The ripple effect reaches beyond Victoria, influencing Digital Twin initiatives and resonating at the national level.

The outcomes of the project will include pilot tools, technologies, and guiding principles for managing satellite-derived, machine-learned data and integrating it into the Vicmap maintenance program. This will enable Vicmap to become an actionable linked data repository, where enriched data can coexist with authoritative data and inform updates.

Variation to the extent of a wetland-swamp area at different time periods. Green polygons illustrate the area extent included in the current Vicmap hydro dataset, while red polygons show the current and actual extent of the area for this hydro feature.

Variation to the extent of a wetland-swamp area at different time periods. Green polygons illustrate the area extent included in the current Vicmap hydro dataset, while red polygons show the current and actual extent of the area for this hydro feature.

By enriching data with contextual information, such as what has changed, how it has changed, and why, Dynamic Vicmap ensures trust in dynamic data, the traceability of updates, fosters user insights, and enables decision-making based on data. Above all, it addresses the challenge of managing increased workflow complexity.

As we anticipate the project’s scheduled completion in mid-2024, the project will demonstrate the boundless potential of spatial innovation when advanced AI, space-based data, and unwavering collaboration come together.

This work has been supported by the SmartSat CRC, whose activities are funded by the Australian Government’s CRC Program, and the Victorian Government, via the Victorian Department of Education and Training.