Wind Turbine Drone Imaging for Precise Autonomous Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for inspecting wind turbines using drones face challenges in generating accurate calibration data for autonomous flight paths due to insufficient or imprecise manufacturer data and lack of reliable models, leading to inefficient and costly manual inspections.
Innovation Solution
A method and system for detecting objects by flying a drone along the wind turbine multiple times from different positions to generate image sets with associated position and location information, recognizing features, and determining their positions using AI or manual evaluation, which allows for the creation of a precise model for autonomous inspection flights.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual inspection methods with climbing teams are used, then inspection quality can be maintained, but safety risks increase and productivity decreases
Solution Approach 1:
The patent replaces manual climbing inspection with an automated drone system that uses optical sensors and image processing algorithms to detect and classify wind turbine components. The drone captures images of the wind turbine from multiple positions, and automated feature recognition identifies blade tips, hub, and other components without human intervention, thereby eliminating safety risks while maintaining inspection quality and increasing productivity
Solution Approach 2:
The system performs self-calibration by automatically detecting geometric features of the wind turbine and computing camera parameters and positioning data without requiring manual intervention. The automated feature recognition and path planning enable the drone to independently execute inspection tasks, reducing dependency on specialized personnel and significantly improving inspection efficiency
2Productivity
If automated drone inspection is implemented, then productivity increases, but measurement precision deteriorates due to insufficient calibration data
Solution Approach 1:
The patent implements a preliminary calibration phase where the drone performs multiple flight passes around the wind turbine to collect image data for computing camera parameters and positioning accuracy. This preliminary action establishes the geometric model and calibration data needed for subsequent automated inspections, ensuring high measurement precision while maintaining productivity
Solution Approach 2:
The system uses feedback from detected geometric features to continuously refine camera parameter estimates and positioning data. The automated feature recognition provides feedback on detection accuracy, which is used to adjust and improve measurement precision in real-time during the inspection process
3Measurement precision
If drones fly closer to wind turbines for better image quality, then measurement precision improves, but collision risk increases
Solution Approach 1:
The patent implements dynamic path planning that automatically adjusts the drone's flight trajectory based on real-time positioning data and detected wind turbine features. The system computes optimal flight paths that maintain the required imaging distance for high-quality images while dynamically avoiding collision risks with moving turbine components, thereby achieving both measurement precision and safety
Solution Approach 2:
The system uses computer vision algorithms and geometric modeling as intermediaries to indirectly measure wind turbine features without requiring the drone to physically approach close to the turbine. By processing images and computing three-dimensional positions through algorithmic reconstruction, the system achieves high measurement precision while maintaining a safe operational distance
4Measurement precision
If multiple image captures from different positions are performed, then measurement precision improves, but loss of time increases
Solution Approach 1:
The patent implements continuous flight paths where the drone captures images at multiple positions along a predetermined trajectory without stopping or repositioning between shots. This continuous action efficiently collects calibration data and feature information from multiple angles, achieving high measurement precision while minimizing the total inspection time
Data Source
AI summary
A method for detecting an object describable by a plurality of predetermined features comprises flying along the object and detecting several portions of the object using at least one recording unit. Each of the portions is detected multiple times from different positions of the recording unit to generate a set of images. Position and location information of the recording unit are associated to each image. Additionally, the method has recognizing features in the image sets and determining the positions and/or locations of the features using the position and location information of the images which contain the features.