The National Research Council Canada is developing an advanced AI system to enable drones (unmanned aerial vehicles) to operate safely in complex environments beyond visual line of sight (BVLOS). This system combines artificial intelligence, computer vision, and radar technology to detect obstacles, avoid collisions, and autonomously navigate around critical infrastructure for inspection and monitoring purposes.
The system uses AI-enabled object detection and threat identification to allow multiple drones to safely perform high-precision inspections, mapping, and automatic condition assessments of essential infrastructure and superstructures. The technology addresses three key challenges: flight path planning and safe landing, real-time detection of objects and potential intruders through onboard sensors, and accurate 3D modeling and photogrammetry.
This system is currently in development and does not involve the collection of personal information. It is designed for use by both National Research Council employees and members of the public who may benefit from drone-based infrastructure inspection services. The system has been tested through over 20 flight sessions in multiple Canadian locations including Ottawa, Montreal, Victoria, and Vancouver Island.
