UK SOVEREIGN AI

Complex signal processing

Mind Foundry’s sensor fusion and complex signal processing engine can handle various signal types, including sonar, radar and cyber. It combines cutting-edge AI with tried and tested signal processing techniques, to process, analyse and enrich feeds from the complete range of modern sensors.

Reduce Cognitive Load

Modern sensor capabilities in radar, sonar, and other modes produce data volumes that outstrip a human's ability to process them. Our solutions help to reduce the load by extracting human-readable outputs from raw data streams.

Increase Autonomy

Uncrewed vehicles provide access to more challenging environments. CSP solutions running at the edge can increase decision-making capability of these platforms when communications are degraded or unavailable.

Do More With Less

Traditional AI architectures have data requirements that scale with a problem's complexity. In complex sensing problems this leads to unrealistic demands for large datasets. Our solutions directly encode expert knowledge to combat this.

Problem complexity and lack of clean data make AI modelling challenging in domains like sonar, radio, and communications. Mind Foundry's MASINT architecture enables us to encode human expertise directly into the model to reduce data requirements and improve accuracy. We work with partners to configure processing pipelines across a range of use cases and interaction modes, including detection, denoising, tracking, and classification.

Deployment agnostic
Signature detection
Signature classification
Signature identification
Search orchestration
Sensor fusion
Exogenous data integration
Explainability views
Environment exploitation

Environmental noise reduction
Local environmental calibration
Edge deployable
Iterative updates
User feedback iteration

More Data, More Problems

Ever-increasing data volumes should lead to improved situational awareness, but only if they are processed and presented intuitively for humans to understand. Instead of finding more knowledge, valuable information is getting lost in all the noise. The nature of the sonar problem space makes it especially challenging from a signal-processing perspective.

Complex Signal Processing

Our sensor integration and processing engine provides foundational composable pipelines for both processing and making inferences on physical sensor data without requiring large datasets and exemplar functional telemetry. Instead, the framework enables us to rapidly onboard new sensors by incorporating information about the sensor and its environment explicitly into the pipeline.

AI Across Sensors

Mind Foundry’s solution provides more flexibility for networked IoT and sensor systems, reduces the need for large datasets to train models on new sensors, and improves the robustness, assurance, and explainability of the systems built within it through increased determinism of pipelines.

5 min read

AI and Sonar: Seeing with Sound

by Nick Sherman

Despite the ocean covering over 70% of the Earth’s surface, we have only physically explored 5% of it. Part of the reason why so much of it is left...

6 min read

Designing Systems with AI in the Loop

by Alistair Garfoot

An AI-in-the-loop approach can help mitigate some of AI’s inherent risks, but to deliver real impact in Defence & National Security, AI must also be...

8 min read

Women in Defence & National Security

by Mind Foundry

Today, women are significantly underrepresented in Defence & National Security. This piece shares insights from some of the women at Mind Foundry...