Skip to main content

The case for a military Spatial Digital Twin

Digital world depicted from lines at front of the image to terrain at the furthest point of the image

The future aiming mark for geospatial intelligence capabilities within Defence should be the creation of a Spatial Digital Twin: a single, dynamic dataset that represents the physical world in sufficient resolution to act as the reference point for all systems requiring mission data. As a phrase du jour there are currently many competing definitions of Digital Twin, but the most relevant to this concept being "a dynamic virtual representation of a physical object or system across its lifecycle, using real-time data to enable understanding, learning and reasoning".[1]

One of the great challenges in achieving a digitised ADF is ensuring the consistency of data (and therefore sensor inputs and outputs) across all platforms and systems. Without consistent data about the battlespace, friendly forces and threats, ‘fifth generation’ forces risk becoming a disrupted network of incompatible information and data that fails to make sense of the battlespace or deliver a warfighting advantage.[2] Indeed, one of the challenges faced with our increasingly advanced systems is the need to provide consistent Intelligence Mission Data to permit these systems to operate in even the most benign environments. Geospatial information underpins all of these functions, and without consistency of information, our forces will not be ‘fighting off the same map’.

While sensors and systems that actively exploit Machine Learning (ML) and limited Artificial Intelligence (AI) functions are already operating in various locations across Defence, it will be some years before we achieve widespread integration of these capabilities. In the intervening years the ADF has a fleeting opportunity to ensure we build a force that is future ready by being AI-ready. Now is the time to get the fundamentals right: exquisite and expensive AI weaponry will be largely redundant if it is operating off different information to the humans it is supporting. A Spatial Digital Twin would provide the foundation of this information by guaranteeing the quality and consistency of the geospatial information these systems consume. Its utility to AI systems would be ensured by possessing sufficient resolution (spatial, spectral, temporal and radiometric) to ensure our human-machine teams are able to fight off the same map. Similar capabilities have recently been applied to civilian infrastructure by Geoscience Australia,  CSIRO’s Data61 / NSW Spatial Digital Twin partnership, United Kingdom’s national Digital Framework Task Group, and Singapore's city planning toolset.

The consistency of information enabled by a Spatial Digital Twin provides several significant benefits for the ADF.

Provides consistency of information

Fundamentally, it provides a set of geospatial information that ensures our forces have access to the most useful and accurate information with a known consistency across the force. Not all the information will be the same; different requirements for the detail of information, formats and standards will persist, but divergence from the common baseline will be more the exception than the norm, and more quantifiable and trackable.

Enables computationally inexpensive AI/ML processes

Once collected and ingested, data storage on tactical platforms is comparatively cheap, but information processing is expensive in time, power and memory. Large datasets can therefore be stored locally for the conduct of change detection conducted closer to the point of detection, which is computationally inexpensive compared to most ML algorithms. By conducting change detection, the locations and events that require investigation by more computationally (and hence bandwidth) expensive algorithms are immediately reduced to only areas where changes or anomalies have been detected.

Supports an AI/ML view of the current battlespace

Machines have difficulty interpreting context and nuance, but excel at the detection of change or anomalous information. By maintaining a detailed and consistent view of the battlespace, changes become almost immediately apparent, allow their detailed interrogation to reveal further information. This supports a more comprehensive understanding of the battlespace—rather than just focusing where a sensor has detected localised change, more global and widespread change detection can occur. Such an approach supports widespread monitoring, cueing humans/machines towards hotspots and trends, and allows the more effective management of ISR at multiple levels.

Recognises the importance of information in battle

This approach treats data like fuel—it is carefully extracted, refined, standardised, and delivered to the user via a carefully managed transport and QA/QC process. By building a trusted repository of data, other information can be compared and contrasted, and considered decisions can be made about its ongoing relevance. This permits the considered ingestion and update of the most useful data, the temporary storage of data that could provide advantage, and the deliberate deletion of data no longer required.

Underpins cooperative targeting and operations in denied or degraded Position, Navigation and Timing (PNT) environments

Correctly correlating what different sensors are targeting is the fundamental requirement of cooperative targeting or engagement, and most systems compare the sensor’s field of view against a geospatial dataset to ensure the consistency of the target. A Digital Twin is thus an essential pre-requisite for cooperative targeting (particularly by human-machine teams) operations in environments where GPS signals and other precision positioning systems are degraded or denied. Such a database is therefore essential given such disruption is expected in high intensity conflict where these collaborative engagement technologies will be most useful, as well as to permit the use of precision weapons. By extension, the Digital Twin data provides extensive mitigations should Position and Navigation functions be disrupted.

Provides consistent data across planning, simulation and rehearsal activities. A single, accurate and detailed representation of the battlespace that is suitable for operational planning (and monitoring, including real-time modifications) and simulation. To capitalise on this information advantage simulation must be expanded beyond training to include rehearsal and probabilistic modelling to refine plans (what we currently call wargaming) and to extend it to determine, for a given range of inputs, which course of action has the greatest chance of success.

The creation of a Spatial Digital Twin is an essential pre-requisite for many of the advanced technologies and capabilities the ADF aspires to master in the era of Accelerated Warfare. The establishment of this capability should be a central focus of Defence’s geospatial and Intelligence, Surveillance and Reconnaissance capability development programs.


[1] Bolton, McColl-Kennedy, Cheung, Gallen, Orsingher, Witell & Zaki, (2018)  

[2] Intelligence Mission Data has five components: Characteristics and Performance, Electronic Warfare Integrated Reprogramming, Signatures, Order of Battle and Geospatial Intelligence.

The views expressed in this article and subsequent comments are those of the author(s) and do not necessarily reflect the official policy or position of the Australian Army, the Department of Defence or the Australian Government.

Using the Contribute page you can either submit an article in response to this or register/login to make comments.