Tunnel UAV Defect Detection With GPS-Free Pose Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional UAV-based tunnel defect detection systems face challenges due to low-quality images in dim tunnels, lack of GPS positioning, and inaccurate pose estimation in highly symmetrical environments, leading to errors in defect localization.
Innovation Solution
Equipping UAVs with LED modules for illumination, integrating IMU, camera, laser radar, and ultrasonic distance meter for pose estimation, and using a trained defect detection model to accurately detect and hover over suspected defects in tunnels without GPS, enhancing image quality and fusing sensor data for precise positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional UAV patrol inspection uses GPS signals for positioning, then real-time location update is achieved, but GPS technology cannot be used in tunnels leading to positioning errors
Solution Approach 1:
The patent introduces an intermediary positioning system that uses the tunnel wall as a reference surface. The UAV captures images of the tunnel wall, and feature points are extracted and matched with a pre-built three-dimensional model of the tunnel to calculate the UAV's position and pose. This intermediary approach (using tunnel wall features as mediators) enables positioning in GPS-denied environments.
Solution Approach 2:
The patent replaces the GPS satellite-based electromagnetic positioning system with a visual-mechanical positioning system. By substituting GPS with a camera-based visual odometry system that processes tunnel wall images and matches them with three-dimensional models, the system achieves positioning functionality adapted to the tunnel environment.
2Ease of operation
If tunnel defect detection is performed in dim light conditions, then UAV can operate in tunnels, but collected images are of low quality
Solution Approach 1:
The patent applies preliminary action by pre-processing captured images through denoising and enhancement algorithms before defect detection. This preliminary processing removes noise introduced by low-light conditions and enhances image quality, making the images suitable for subsequent defect detection tasks.
Solution Approach 2:
The patent changes image parameters through digital processing. By adjusting brightness, contrast, and applying denoising filters, the system transforms low-quality dim-light images into enhanced images with improved quality metrics, enabling effective defect detection despite poor lighting conditions.
3Device complexity
If UAV uses simple positioning methods in tunnels, then system complexity is reduced, but pose estimation becomes inaccurate due to symmetrical structure
Solution Approach 1:
The patent transitions from two-dimensional image features to three-dimensional pose estimation. By matching two-dimensional extracted features with a three-dimensional tunnel model, the system calculates comprehensive pose information including position (x, y, z) and orientation (roll, pitch, yaw). This dimensional transformation enables accurate pose estimation despite tunnel symmetry.
Solution Approach 2:
The patent addresses tunnel symmetry by introducing asymmetric reference frames and coordinate systems. By establishing a global coordinate system and using asymmetric feature point matching strategies, the system distinguishes between symmetric tunnel sections and achieves unique pose estimation even in highly symmetrical environments.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate defect detection and hovering in tunnels with dim lighting and symmetrical structures, improving image quality and positioning accuracy, and enhancing detection robustness through data enhancement and sensor fusion.
Implementation Method 1
The UAV according to the invention uses LED (light-emitting diode) module to supplement the illumination in the tunnel
Implementation Method 2
integrates IMU (Inertial Measurement Unit), camera, laser radar and ultrasonic distance meter to realize the pose estimation of the UAV
Implementation Method 3
integrates IMU (Inertial Measurement Unit), camera, laser radar and ultrasonic distance meter to realize the pose estimation of the UAV
Data Source
AI summary
Tunnel defect detecting method and system using unmanned aerial vehicle (UAV) are provided, and the UAV is equipped with a light-emitting diode (LED) module, a camera, a laser radar, an ultrasonic distance meter and an inertial measurement unit (IMU). The method includes: collecting images in a tunnel based on the LED module and the camera to obtain a training image set; training by using the training image set to obtain a defect detecting model, collecting real-time tunnel images, detecting suspected defects to the real-time tunnel images by the defect detecting model, obtaining pose information of the UAV based on the camera, the laser radar, the ultrasonic distance meter and the IMU to control the UAV to hover. The method can realize accurate pose estimation and defect detection in the tunnel with no GPS signals and highly symmetrical inside.