PTZ Camera ROV Localization Near Nuclear Reactors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In nuclear reactor inspection environments, existing sensor modalities like inertial measurement units, gyroscopes, and GPS are constrained by radiation, making it difficult to achieve accurate localization of underwater robots, while vision-based systems are viable due to low turbidity, but require robust automation for reliable operation.
Innovation Solution
A tightly integrated system combining computer vision and state estimation algorithms with an underwater pan-tilt-zoom (PTZ) camera, utilizing variable focal length to mitigate radiation exposure and enable automated localization of ROVs, including self-initialization, self-calibration, and automated failure detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If radiation-hardened electronics are used near reactor walls, then reliability is improved, but measurement precision deteriorates due to image noise
Solution Approach 1:
The patent introduces an external camera positioned away from the radiation source as an intermediary device. This camera captures images of the ROV and vessel wall without being directly exposed to high radiation levels, thereby avoiding the image noise and degradation that plague in-situ radiation-hardened sensors while still enabling localization functionality.
Solution Approach 2:
The system transitions from direct 3D localization near the radiation source to indirect 2D image-based localization from a remote position. By capturing 2D images from a safe distance and using computer vision algorithms to extract positional information, the system achieves localization without exposing sensors to the harmful radiation environment.
2Manufacturing precision
If the camera is positioned close to the vessel wall for inspection, then inspection quality is improved, but radiation exposure increases causing sensor degradation
Solution Approach 1:
The external camera acts as a mediator that enables inspection functionality without being exposed to radiation. The camera is positioned outside the immediate radiation zone but can still capture images of the vessel wall and ROV, allowing inspection operations to proceed without compromising sensor integrity.
Solution Approach 2:
The patent replaces direct mechanical/optical sensing near the radiation source with remote visual sensing. Instead of using sensors that physically interact with the radiation environment, the system uses an external camera that observes the scene optically from a protected position, substituting direct measurement with indirect observation.
3Ease of operation
If automated localization algorithms are implemented, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The localization system is designed to be self-sufficient, automatically extracting features from images, tracking the ROV, and computing position and orientation without requiring manual intervention. The system performs self-initialization and continuous self-calibration, eliminating the need for operators to manually configure or adjust parameters while maintaining operational simplicity.
Solution Approach 2:
The system performs preliminary actions such as self-initialization and self-calibration using pre-stored vessel geometry data before actual localization begins. By preparing the system in advance with known structural information, the algorithms can quickly and automatically determine ROV position without requiring complex real-time manual setup or calibration procedures.
Data Source
AI summary
A vision-based state estimation framework to estimate the state of an underwater remotely operated vehicle (ROV) used for inspection of an underwater structure, for example, a nuclear reactor pressure vessel. The framework employs an external overhead, pan-tilt-zoom (PTZ) camera as the primary sensing modality and incorporates prior knowledge of the geometry of the structure.