ROV Video Localization Using Fiducial Markers in Reactor Vessels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for locating remotely operated vehicles (ROVs) in inhospitable environments like nuclear reactors are either costly or ineffective due to radiation interference, and existing solutions require sensors on the ROV or expensive equipment like lasers.
Innovation Solution
A method using a video camera to process video feeds, identify landmarks, and determine the position of fiducial markers on the ROV within a calibrated three-dimensional framework, allowing for precise location determination without sensors on the ROV or costly equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are installed on the ROV for location tracking, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent uses visual copying by capturing images of fiducial markers on the ROV with an external camera system. Instead of using sensors on the ROV, the system creates a visual copy of the marker positions in the workspace environment, processes these images to determine ROV location, thereby achieving accurate tracking without adding complex sensing equipment to the ROV itself
Solution Approach 2:
The patent introduces fiducial markers as an intermediary element between the ROV and the camera system. These markers serve as a mediator that carries positional information visible to the external camera, allowing the system to track ROV location indirectly through marker detection rather than direct sensor measurement, simplifying the overall system architecture
2Measurement precision
If lasers or expensive equipment are used for location determination, then measurement precision is improved, but cost increases
Solution Approach 1:
The patent employs inexpensive fiducial markers (such as colored stickers or printed patterns) instead of expensive laser systems or specialized tracking equipment. These markers are simple, low-cost visual elements that can be easily manufactured and attached to the ROV, providing sufficient positional information when captured by standard digital cameras without requiring costly specialized hardware
Solution Approach 2:
The patent replaces mechanical/optical systems like lasers with a vision-based system using standard digital cameras and image processing. Instead of using active laser emitters and receivers, the system uses passive visual detection of markers through camera imaging and computational algorithms, substituting expensive optical-mechanical systems with more affordable electronic imaging and software processing
3Reliability
If radiation-resistant sensors are used on the ROV, then reliability in radiation environments is improved, but device complexity and cost increase
Solution Approach 1:
The patent extracts the sensing function from the ROV itself and relocates it to the external environment. By placing the camera system outside the radiation zone and using passive fiducial markers on the ROV that don't require active electronic sensors, the system eliminates the need for radiation-resistant sensors on the ROV, thereby maintaining reliability in radiation environments while reducing system complexity
Solution Approach 2:
The fiducial markers on the ROV are passive visual elements that reflect or emit visible light without requiring active electronic components or power sources on the ROV. The markers essentially serve themselves by being visible to the external camera, eliminating the need for complex radiation-hardened sensor systems while maintaining tracking functionality in radiation environments
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention provides a method and system for locating a remotely operated vehicle within a three-dimensional workspace. The method comprises: receiving a video feed of the workspace from a video camera; determining the location in the video feed of a number of fiducial markers on the remotely operated vehicle operable to travel in three dimensions within the workspace comprising an interior of a nuclear reactor vessel; determining a three-dimensional position of the remotely operated vehicle within the workspace using the location of the number of fiducial markers in the video feed; processing the video feed to identify landmarks, and features, thereof of known physical structures in or near the workspace; determining a correlation between the landmarks and the features identified in the video feed and known physical structures, comprising constructing a three-dimensional reference framework based on previously known dimensional information of the known physical structures; and calibrating the video feed from the video camera to the known physical structures using the correlation.