URBANFLUX
Non-cooperative drone detection · working prototype

Grow the signal.Detect the anomaly.

UrbanFlux helps detect non-cooperative drones by combining publicly available signals and privacy-preserving aggregate data into a shared, explainable picture, so anomalies can be corroborated before they become operational surprises.

Designed for situational awareness without identity-based surveillance or storage of personal imagery.

Working prototypeValidated in simulationField validation next
Validation plan

Four phases. Each closes on a measurable criterion.

A twelve-month programme taking the system from a simulation-validated prototype to capability proven in the field. Every phase ends on a criterion that can be checked independently.

TRL 6 to TRL 7

  1. 01Months 1 to 3

    Live sensor integration and data quality

    Closes when

    Two or more heterogeneous live feeds sustain continuous ingest and survive the loss of any one source. Signal-to-noise and end-to-end latency are characterised against a threshold agreed with the technical authority.

  2. 02Months 4 to 6

    Detection trials

    Closes when

    Detection rate, false-alarm rate and confidence calibration are measured against two drone types with distinct flight signatures. Passive RF sensing is brought online against live signals.

  3. 03Months 7 to 9

    Edge deployment and interoperability

    Closes when

    An edge node runs in the field, emitting CoT and TAK to a partner command system, with behaviour tested under degraded conditions. A privacy impact assessment confirms the pipeline retains anonymised aggregates only.

  4. 04Months 10 to 12

    Generalisation

    Closes when

    The same core, unmodified, extends to a second threat class and a second environment through connectors alone.