Outpacing the Drone Threat Through Bidirectional Live-Virtual-Constructive Training

12 May 2027
New Horizons Keynote Theatre
Outpacing Autonomous and Uncrewed System Threats

The rapid evolution of small uncrewed aerial systems presents a growing training challenge.
Personnel must learn to detect, track and respond to agile aerial threats whose tactics and
behaviours can evolve faster than conventional training solutions. However, representative
counter-UAS training is constrained by safety, cost, airspace and equipment limitations. Flying
drones close to trainees can introduce unacceptable risk, while repeatedly damaging physical
targets is neither economical nor sustainable. Fully synthetic scenarios remove these
constraints, but they may not reproduce the unpredictability, adaptation and decision-making
of a human-operated drone.
This paper presents a modular Live-Virtual-Constructive (LVC) training concept that combines
real-world drone activity, human-operated simulators and computer-controlled entities in a
shared counter-UAS training environment. In its simplest configuration, a live operator flies a
drone in a controlled physical area, a camera-based detector tracks the drone's position in near
real time. The detected drone's 3D position is projected into the virtual environment, where a
trainee encounters a corresponding digital target.
The trainee can practise detection, tracking, identification and simulated engagement without
applying any physical effect to the aircraft. Information about the trainee's actions can also be
returned to the drone operator, allowing the pilot to adapt the flight profile in response to virtual
detection or engagement. This bidirectional exchange creates an interactive scenario in which
the actions of both users affect its progression and outcome.
The constructive layer populates the same environment with entities controlled by algorithms or
artificial intelligence. These may include drones following scripted or adaptive flight routes,
simulated drones, detectors and other sensors, and background air traffic. Constructive entities
increase scenario density and complexity beyond what a small number of live and virtual
participants can provide, allowing a single human-flown drone to be embedded within a larger
synthetic threat picture or a coordinated swarm.
When live flying is impractical, the physical drone can be replaced by a human-operated drone
flight simulator while retaining the same data interface. Live, virtual and constructive inputs can
be mixed or substituted, enabling progressive training, repeatable scenarios and
experimentation with threat behaviours that would be unsafe or difficult to reproduce
physically.
The proposed proof of concept evaluates whether a low-cost sensing and data-exchange
architecture can provide sufficient positional accuracy, update rate and end-to-end latency for
credible training. It also examines coordinate alignment, bidirectional interoperability, and the
consistent representation of live, virtual and constructive entities within a common operational
picture. The work aims to identify the minimum technical conditions required for an effective
LVC counter-UAS exercise. By combining authentic human-controlled threat behaviour with
scalable computer-generated forces and virtual training effects, the concept could help forces
rehearse emerging drone threats earlier, more frequently and with lower risk.

Chairperson
David Elsey
David Elsey, Analysis and Design Portfolio Technical Authority - Training & Simulation Group - BAE Systems
Speakers
Aapo Nikkola
Aapo Nikkola, Project Advisor - Turku University of Applied Sciences
Timo Haavisto
Timo Haavisto, Senior Advisor - Turku University of Applied Sciences