3D Sensor Registration Using Tagged Reference Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional guarding technologies, particularly those using 3D sensors, face challenges in configuring and registering systems to ensure human safety around industrial machinery, requiring advanced skill sets and significant overhead, and are inflexible for collaborative human-machine tasks.
Innovation Solution
A method for registering 3D sensors using computer-vision techniques and registration objects with distinctive signatures, allowing automatic and efficient alignment of sensors to each other and to machinery, with continuous monitoring for real-time safety assurance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional guarding technologies use 3D sensors for monitoring workspace, then safety monitoring capability is improved, but system complexity and configuration difficulty increase
Solution Approach 1:
The patent introduces registration objects as intermediaries between the 3D sensors and the machinery. These objects with distinctive features serve as mediators to establish coordinate transformations, simplifying the configuration process by providing known reference points that automatically define spatial relationships without requiring complex manual calibration procedures.
Solution Approach 2:
The patent implements preliminary registration of sensors to registration objects before actual safety monitoring begins. By pre-establishing coordinate transformations through registration objects that are placed in known positions, the system prepares the spatial reference framework in advance, eliminating the need for complex real-time calibration during operation.
2Adaptability or versatility
If manual configuration methods are used for sensor registration, then flexibility is maintained, but time consumption and skill requirements increase
Solution Approach 1:
The patent enables the system to perform self-registration by automatically detecting registration objects in the workspace and computing coordinate transformations without human intervention. The sensors autonomously identify the distinctive features of registration objects and calculate their spatial relationships, eliminating the need for manual configuration while maintaining adaptability to different workspace layouts.
Solution Approach 2:
The patent replaces manual mechanical configuration processes with automated computer vision-based detection and computational geometry methods. Instead of requiring operators to physically measure and input coordinate data, the system uses image processing to automatically detect registration object features and compute transformations, significantly reducing time and skill requirements.
3Productivity
If tight coupling between machine and human actions is achieved using 3D sensors, then productivity is improved, but false positives increase safety shutdowns
Solution Approach 1:
The patent implements multi-functional safety zones that can dynamically adapt their boundaries and monitoring parameters based on the actual task being performed. The same sensor system serves multiple functions: monitoring intrusion, tracking human position, and adjusting safety parameters according to machine operation state, thereby reducing false positives while maintaining tight coupling for productivity.
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 flexible and efficient monitoring of workspaces for human safety, reducing processing time and complexity, and ensuring reliable operation of industrial machinery by transitioning to a safe state upon registration deviations or signal degradation.
Implementation Method 1
3D depth sensors have been recently employed in various machine-guarding applications. Examples of the 3D depth sensors include 3D time-of-flight cameras, 3D LIDAR
Implementation Method 2
two-dimensional (2D) LIDAR sensors that use active optical sensing to detect the minimum distance to an obstacle along a series of rays emanating from the sensors
Data Source
AI summary
Image sensors distributed about a workcell including industrial machinery are registered using a registration object and an information tag associated therewith. The tag contains information specifying the location of the object and/or the pose of the object. This information is acquired along with images of the registration object, and the sensors are registered based at least in part on the images and the acquired information.