The next step - AI-supported Battlefield Management Systems (II)

Part II - Technical Requirements and Real-Life Scenarios

Part II - Technical Requirements and Real-Life Scenarios

Welcome back to the Resilience Newsletter!

In today’s edition, we continue our three-part series on the integration of Artificial Intelligence (AI) into Battlefield Management Systems (BMSs). Last time, we examined the limitations of existing systems and explored the foundational potential of AI to enhance battlefield decision-making and operational efficiency.

In this part, we will explore which available AI products are suitable for integration and how such a fusion could look in a combat scenario. As in the previous edition, we are pleased to be guided once again by our guest author Linus Wittenburg. Linus is completing his MSc in Management at the London Business School and has worked in tech consulting for several years, during which he gained a specialisation in AI applications. He is also a 2nd lieutenant in the German Armed Forces, specialised in Reconnaissance.

Yours,
Uwe, Jack and Jannic

The AI Integration Process

As we have learned in Part I, AI can enhance decision-making, situational awareness, and operational efficiency by leveraging the hardware and software capabilities of Battlefield Management Systems (BMSs). However, integrating such models requires a systematic approach to ensure seamless functionality and compatibility with hardware units. The following section provides an overview of the implementation process:

Model Selection and Optimisation 

Battlefield operations come with unique challenges, including rapidly changing environments, high-stakes decision-making, and the need to process diverse data sources in real-time. While such requirements call for specialised software, there is a variety of pre-trained AI models in the civilian sector that can be fine-tuned to meet individual demands. Examples include the AI in Skydio’s X10 drones, which automatically recognises, documents, and analyses its surroundings based on objectives. Similarly, the ArcGIS smart model can create and share interactive 3D maps of the terrain.[1] [2] These models can be fine-tuned using mission-specific data to fit the needs of military tasks. Additional adjustments such as an optimisation for edge computing ensure that the AI can run smoothly on portable BMS terminals.

Fig. 1: A Skydio drone using AI to inspect a bridge for damage.
Source: Skydio

Deployment and Integration

After the AI is trained, it’s packaged into a container using specialised tools to run smoothly anywhere. This ensures compatibility across diverse environments while simplifying deployment and updates. Additionally, this allows the model to process real-time data from sensors or any other source. 

The container can then be integrated into the BMS architecture, connecting data inputs such as troop positions or satellite feeds with the AI processing engine. This is possible due to the multi-platform capabilities and powerful CPUs of modern BMS hardware, allowing seamless integration of various operating software and AI accelerators. To further enhance performance, special toolkits can be used at this point. [3]

Testing and Scalability

Following its integration, rigorous testing ensures the AI’s reliability under operational conditions. This ensures that the tool performs as intended in various scenarios while aiding the localisation and elimination of potential weaknesses. Once the model has passed all performance tests, the AI can be integrated into existing military structures and, if applicable, scaled up to a multinational collaborative level. The latter approach will be discussed in more detail in part III of the series.

Intelligent BMS in Use - What AI-Aided Combat Scenarios Could Look Like

Its ability to automatically infer priorities and recommendations from new data points provides commanders with an information advantage during missions. By processing and analysing battlefield data in real-time, the intelligent BMS transforms raw information into actionable intelligence that can significantly impact operational outcomes.

To illustrate how such a system functions in practice, let’s examine a realistic combat scenario where an intelligent BMS could analyse, adapt, and respond to rapidly evolving battlefield conditions.

1. The Combat Situation

Blue Forces have established a defensive posture marked by the Forward Line of Troops (FLOT), with key units strategically positioned to guard against a frontal assault. Despite these preparations, vulnerabilities exist, particularly in the southern sector, where red forces have identified weaknesses caused by overstretched defensive lines. 

Recognising this opportunity, red forces plan to exploit this gap and are now mobilising for a surprise breach of the FLOT. Shortly before the surprise attack, a UAV captures imagery of the advancing enemy forces.

Fig. 2: Tactical overview of incoming attack.
Source: Linus Wittenburg

2. The AI Response

As soon as the AI-supported BMS receives the visual information from the UAV, the system will autonomously initiate the following responses:

Threat Detection and Analysis

The AI-aided BMS processes the surveillance drone data and identifies the incoming armoured forces on the flank. It assesses the threat’s size, speed, and weaponry and classifies its severity. Simultaneously, it predicts the time until contact with friendly forces and assesses how this may impact the ongoing mission.

A flashing warning (“Incoming Attack Detected”) is displayed to all units in the area of interest. Additionally, the BMS overlays risk zones on the digital map, visually marking the threat’s trajectory and highlighting vulnerable sectors. Command automatically receives a range of recommendations along with follow-up actions, enabling immediate and effective response measures. (Fig. 3)

Real-Time Adjustments and Troop Coordination

Based on the current situation picture, the BMS dynamically updates troop positioning recommendations, recommending immediate redeployment of available reserve units to strengthen the defensive line. Meanwhile, the system maintains communication and data synchronisation across allied units, including detailed instructions to the southern forward command post to prepare a counterattack.

Resource Allocation and Artillery Support

The BMS identifies which units are most critical for the countermeasure based on ammunition types and available resources. It then prevents potential shortages by proactively assigning supply runs to these positions before the shortages can occur.

In order to delay the advance, the system recommends artillery coordinates for preemptive strikes and requests close air support, if available. The proposed strike zones are mapped out and integrated into the system's situational view. In addition, the most likely enemy fallback routes are blocked using scatter mines.

Individual Commands

Besides large-scale predictions and recommendations, the AI-supported BMS assists individual units in their defence of positions. Interactive map layers show individual troops’ current positions and suggested relocations based on terrain, weaponry and enemy movement. This ensures that squad leaders can position their units in the most effective defensive positions. To minimise distractions, tailored pop-up notifications deliver concise orders to respective commanders, including suggested paths and observation corridors.

Impact Prediction and Contingency Planning

Throughout the commencing engagement, the BMS actively reevaluates key metrics and adjusts recommendations. A mission summary panel tracks key data such as expected outcomes, estimated delays to enemy progress, and resource availability, relaying major changes directly to higher command. 

Besides the high-level overview of the current situation, if the collapse of the defensive line is considered imminent, the BMS also prepares to provide fallback routes. 

Assisted by the real-time assessment and response capabilities of the AI-supported BMS, blue forces are able to delay the enemy push through a combination of rapid repositioning and targeted artillery support. This gives reinforcements time to move up to the FLOT and stall any further advance of red forces. Following the hammer and anvil tactic, a rapid counterattack is initiated by units stationed in the south, targeting the enemy’s flank. With attacks from two sides and artillery having deployed a scatterable minefield, the enemy forces are pinned down and eventually defeated. 

The breakthrough has failed. The integrity of the defensive line has been preserved.

Fig. 3: Tactical view of AI-supported counter measures.
Source: Linus Wittenburg

This example, although of theoretical nature, highlights how crucial AI could become in determining the outcome of future similar scenarios. Its ability to rapidly process information and infer direct actions from them will become an integral part of future operations, ensuring a faster, more precise execution on an individual level as well as across large-scale engagements. 

In the next edition of the “European Resilience Newsletter”, we will talk about the concept of scalability. Through the implementation of an AI-centered mission approach in multinational operations, new capabilities arise with regards to current conflicts such as the Russo-Ukrainian war and the rising threat of sabotage in the Baltic Sea.

Stay tuned for part 3!

Sources and further reading

News That Caught Our Attention 👀

Every week we feature a list of interesting roles in European DefenceTech start-ups and scale-ups for readers seeking their next challenge in their careers.

If you are a founder and would like to promote your open roles, please get in touch with us!

Passionate and want to contribute? 👩🏻‍💻

The European Resilience Tech Newsletter is always looking for regular and guest authors, writers, reporters, content creators etc. If you like what you read, you are passionate about improving European resilience regardless of your background and want to contribute, just reach out to us!

European Resilience Tech Newsletter Team

Uwe Horstmann co-founded Project A Ventures in 2012 as General Partner and has built Project A to be a leading European early-stage investor with over $1bn USD under management and having backed 100+ founders. In addition to Project A, Uwe serves as Reserve Officer in the German armed forces and advises the German Ministry of Defence in digital transformation issues.

Jack Wang is a software engineer turned product-driven tech investor and joined Project A in 2021 to lead the firm’s deep tech investing, which has grown to include DefenceTech. Prior to joining Project A, Jack worked in a variety of organisations such as Amazon and Macquarie Group across Australia, US and UK / Europe. Jack holds a MBA from London Business School and Bachelors of Engineering (Bioinformatics, 1st) from UNSW, Australia.

Jannic Meyer joined Project A initially contributing to what is now known as the Project A Studio, partnering with founders at the pre-idea stage, where he covered a variety of topics ranging from energy infrastructure to dual-use robotics and led our investment in ARX Robotics. He is now part of the investment team at Project A covering all things resilience.

Project A Ventures is one of the leading early-stage tech investors in Europe with offices in Berlin and London. In addition to 1 billion USD assets under management, Project A supports its 100+ portfolio companies with a platform team over 140 functional experts in key areas such as software and product development, business intelligence, brand, design, marketing, sales and recruiting. Project A have backed founders of Trade Republic, WorldRemit, Sennder, KRY, Spryker, Catawiki, Unmind and Voi as well as founders building in European Resilience: