TRLUP – AI-Powered Drones for Infrastructure and Industrial Inspection

Aeronovous delivers AI-powered drone inspection solutions for infrastructure, industrial equipment, and wind turbines, with this internship specifically focused on wind turbine applications. The traditional inspection industry relies on costly and inefficient manual methods—rope-access technicians, prolonged turbine outages, and highly variable reports influenced by crew experience and weather conditions—creating significant safety risks, operational downtime, and inconsistent defect detection. Even when drones are deployed, human operators must manually review thousands of images to identify defects, a tedious and error-prone process that can allow critical issues to go unnoticed, leading to equipment degradation and expensive failures. Aeronovous differentiates itself through automated defect detection powered by computer vision, deploying a drone workflow that captures standardized imagery of turbine blades, towers, hubs, and nacelles, then automatically detects, measures, and tracks defects, reducing days of manual analysis to hours of automated processing and enabling operators to schedule repairs with minimized downtime, cost, and risk. Currently in a pre-commercial stage, the core technology has been validated in simulation environments but awaits real-world validation with paying customers in operational settings. To achieve commercialization, the company must establish reliable sensor integration, develop repeatable capture protocols for tall structures, generate labeled datasets tailored to client needs, train models meeting accuracy benchmarks under field conditions, and ensure compliance with Transport Canada regulations and site safety procedures—all while demonstrating measurable return on investment to secure early adopters for pilot programs. This project directly tackles these commercialization barriers through market validation and customer discovery, defining turbine inspection requirements and safety protocols, standardizing image capture workflows using the RGB FPV drone platform, establishing infrastructure for data storage and labeling, training and benchmarking defect-detection models, developing standard operating procedures and ROI frameworks, and securing pilot agreements with prospective clients to validate the technology in live operational environments.

Faculty Supervisor:

Sheri Williams

Student:

Partner:

Springboard Atlantic Inc.

Discipline:

Engineering

Sector:

Aerospace; Artificial Intelligence; Green/Alternative Energy

University:

Nova Scotia Community College

Program: