Defence Autonomy · GPS-Denied

Self-Driving Without Sensors in GPS-Denied Zones

The UK MOD needed autonomous vehicles for shooting practice, but the equipment was too expensive to be shot at. We built a radar-based system that tracks and navigates vehicles with sub-meter accuracy — without a single sensor on the car.

Client
UK MOD
Impact
Sub-meter Accuracy
Start your project
Autonomous Humvee being tracked by external radar system for UK MOD target practice
Positioning Accuracy
Sub-meter
beating standard GPS
Sub-meter
Position accuracy
beating standard GPS
0
Recorded crashes
perfect safety record
GPS-denied
Navigation
operates without satellite signal
100%
Flexible system
easily integrates new features
The Challenge

High-performance autonomy for disposable targets

Our team was approached by the UK MOD with a complex problem: they needed autonomous vehicles to practice shooting against moving targets, but standard autonomous equipment was too expensive to be destroyed. Additionally, the vehicles needed to operate in a completely GPS-denied environment.

Convert legacy vehicles
The UK MOD needed to convert old Humvees into autonomous targets for shooting practice.
Extreme cost constraints
Standard autonomous sensors (LiDAR, GPS) were too expensive to be placed on vehicles that would be shot at.
GPS-denied environment
The vehicles needed to operate in areas where satellite navigation was unavailable or unreliable.
Dynamic obstacle avoidance
The system had to identify and avoid moving obstacles appearing in the vehicle's path with low latency.
The Solution

External Radar Sensing — Autonomy without the cost

Our solution was to place radars along the path to detect the vehicle and its surroundings, rather than detecting the environment from the car. We designed a processing pipeline to detect the vehicle with sub-meter accuracy, solving the GPS-denied environment problem while keeping the expensive hardware safe.

Innovative external sensing architecture: radars placed along the path instead of on the car

Bespoke processing pipeline to detect and track vehicles with sub-meter accuracy

Radar-based surround-view system for comprehensive obstacle detection and avoidance

Cost-effective design that keeps expensive equipment safe from direct fire

Robust navigation algorithms that function independently of GPS signals

Real-time processing and control loop for high-speed target maneuvering

The Results

Strategic impact quantified — Sub-meter precision

We were able to meet these challenging requirements in a cost-effective and reliable manner. The system beat standard GPS accuracy using only external radars and hasn't registered a single crash during its operational life.

Sub-meter
Accuracy achieved
Our radar-based system exceeded standard GPS accuracy, providing precise positioning in a GPS-denied environment.
Zero
Collisions recorded
The obstacle avoidance system has proven perfectly reliable, with no crashes recorded throughout the testing phase.
Cost-safe
Infrastructure protection
By moving expensive sensors off-vehicle, we met the requirement for a disposable target system without sacrificing performance.
100%
System flexibility
The modular design allows for easy integration of new features and adaptation to different vehicle types.
"Our team delivered a successful project by placing radars along the path to detect the vehicle and its surroundings, solving the GPS-denied environment problem in a cost-effective and reliable manner."
GI
Gradient Insight
Autonomy Lead
Technology Stack
Radar NavigationGPS-Denied AutonomyObstacle AvoidanceTarget TrackingLow-latency ControlExternal Sensor FusionBespoke AI PipelineDefence Systems AI
At a Glance
Client UK MOD
Industry Defence / Autonomy
Delivery GPS-Denied Navigation
Accuracy Sub-meter validated
Sensing Path-side Radar

Similar challenge?

We build bespoke AI systems for complex, high-stakes domains — from day one.

Book a free discovery call
Ready to build?

Your domain, your data — built to your exact requirements

We don't adapt generic models. We architect bespoke AI systems that meet your performance, licensing, and infrastructure constraints — from scratch.