Skip to content
This is NOT an official site of the Government of Canada. Click here for the official AI registry.

Helpful Places

Government of Canada – DTPR AI Register

AI-Powered Drone Detection and Collision Avoidance

Research & Development · Safety & Security

What it collects

About a measurement
Anonymized data
  • Sensor data captured by on-board electro-optic cameras and radar systems mounted on UAVs, including imagery, range measurements, and object signatures used for detection and collision avoidance.
About a place
Anonymized data
  • Spatial data describing the environment around critical infrastructure and superstructure targets, including aerial imagery and 3D map data gathered during BVLOS flight test sessions in Ottawa, Montreal, Victoria, and Cowichan Territory.
Operational data
Anonymized data
  • Synthetic data generated using in-house UAV simulation software, open-source, and commercial scientific software, supplementing real-world flight test data collected across over 20 UAV flight sessions.
Run by
National Research Council Canada (NRC)
Where
No fixed location
Kept
Not stated by the Helpful Places.
Shared with
Accountable organization
Your copy
You cannot see the data it holds about you. What you can do

What it is for

Research & DevelopmentSafety & Security

This system uses artificial intelligence to enable drones to detect objects and avoid collisions while operating beyond the visual line of sight of a human operator. Developed by the National Research Council Canada, it supports inspection, monitoring, and 3D modelling of critical infrastructure such as bridges and buildings. The system is currently in development and does not collect personal information about members of the public.

What it collects and what happens to it

Data taken in

About a measurement
Anonymized data
  • Sensor data captured by on-board electro-optic cameras and radar systems mounted on UAVs, including imagery, range measurements, and object signatures used for detection and collision avoidance.
About a place
Anonymized data
  • Spatial data describing the environment around critical infrastructure and superstructure targets, including aerial imagery and 3D map data gathered during BVLOS flight test sessions in Ottawa, Montreal, Victoria, and Cowichan Territory.
Operational data
Anonymized data
  • Synthetic data generated using in-house UAV simulation software, open-source, and commercial scientific software, supplementing real-world flight test data collected across over 20 UAV flight sessions.

Processing

Computer Vision
  • AI-enabled computer vision processes electro-optic camera imagery to detect aerial objects, intruders, and infrastructure features during BVLOS drone flight, enabling collision avoidance and photogrammetric 3D modelling.
Classification & Prediction
  • Machine learning models classify detected objects (e.g., threat vs. non-threat) and predict collision risk to support autonomous flight path planning and safe landing decisions during BVLOS operations.

What it does

Sensing (Perceptive AI)
Autonomous
  • The system uses on-board electro-optic and radar sensors to detect objects and intruders in real time, converting raw sensor signals into structured detections that feed the collision avoidance and flight path planning subsystems.
Deciding (Analytical AI)
Autonomous
  • The system classifies detected objects, scores collision risk, and ranks flight path options to enable autonomous BVLOS navigation and safe landing decisions without real-time human review of each determination.

Outputs

About a place
Anonymized data
  • High-precision 3D maps, photogrammetric models, and condition assessment outputs derived from aerial inspection of critical infrastructure and superstructures during BVLOS drone operations.
A physical action
Anonymized data
  • Autonomous collision avoidance manoeuvres and flight path adjustments commanded by the AI system to the drone's flight control systems during BVLOS navigation, including safe landing directives.
A recommendation or prediction
Anonymized data
  • Automated condition assessments and inspection reports generated from 3D models and photogrammetric data, providing recommendations about the state of inspected critical infrastructure.

Run by

National Research Council Canada (NRC)
  • The National Research Council Canada (NRC) is the Government of Canada's primary federal research and technology organisation. NRC is responsible for developing and deploying this AI-powered drone system for aerial object detection and collision avoidance.

Government of Canada AI Register — Department field

Built by

Not stated by the Helpful Places.

Kept for

Not stated by the Helpful Places.

Shared with

Not available to me
  • The system does not involve personal information and does not produce data outputs accessible to individual members of the public. Outputs are operational and infrastructure-related data used by NRC researchers and authorised operators.
Available to the accountable organization
  • Sensor data, flight logs, 3D models, inspection outputs, and condition assessments are available to National Research Council Canada researchers and authorised project personnel.

Stored

Not stated by the Helpful Places.

How to read the colours

Can it identify you?

Anonymized data
Data about people with the link to who is broken. Stripped of identifiers, blurred, aggregated, or noised so this system can’t reasonably tie a record back to an individual.
Pseudonymous data
Each person’s data is tied to a token (hash, ID, template) that lets this system recognise the same person across events, but the token itself doesn’t reveal a name. Reidentification is possible with extra information.
Identifiable data
The data either contains a direct identifier (name, address, account name, recognisable face or voice, plate number) or carries a token this system uses to look up legal identity during processing.

Who completes the loop?

Human decides
This mode suggests; a person decides what to do next. The AI is always advisory — a human is in the loop on every decision. Example: a triage tool ranks cases for a clinician who chooses which to see first.
Human executes
This mode decides; a person carries out the result. Example: an optimizer plans the day’s trash-collection routes, and drivers run them.
Autonomous
This mode decides and acts on its own. No person reviews each decision or carries out the resulting action.

Definitions from the DTPR standard. Amber is about your data, violet about who decides. The fuller the shape and the deeper the colour, the more identifying the data or the less a person is involved.

What you can do

Ask about this system

Questions go to the Helpful Places, not the vendor.

Your rights

  • Right to Algorithmic TransparencyThis system is developed by the National Research Council Canada and is listed on the Government of Canada's AI and Data Solutions Register. Information about the system's purpose, capabilities, and data use is publicly available through that register. The system is currently in development and does not involve personal information.

Risks and safeguards

  • Physical harmAutonomous drones operating BVLOS in complex airspace carry risk of mid-air collision with other aircraft or objects, and risk of crash causing property damage or injury to persons on the ground. Mitigations include the AI-powered threat detection and collision avoidance system itself, extensive real-world flight testing across multiple locations, use of both on-board electro-optic and radar sensors for redundancy, and the system's current in-development status implying controlled testing conditions before operational deployment.