3D ToF Camera Depth Disambiguation for Safety-Critical Applications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D ToF cameras are not safety-rated and cannot be used in safety-critical applications like machine guarding or collaborative robotics due to issues with pixel-level errors and the need for high reliability and accuracy.
Innovation Solution
The proposed system uses a camera architecture with redundant 3D sensors and a controller that disambiguates distances, resolves errors due to periodic range ambiguity, and determines corrected pixelwise values using multiple frequency measurements and stereo vision techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If 3D ToF cameras are used for machine guarding, then 3D depth information is provided, but pixel-level errors and lack of safety rating prevent use in safety-critical applications
Solution Approach 1:
The patent segments the depth measurement task by dividing the image into multiple depth bins and processing each bin independently. This allows identification and isolation of erroneous depth measurements within specific depth ranges, preventing them from compromising overall system reliability while maintaining measurement precision in valid ranges.
Solution Approach 2:
The patent implements feedback mechanisms where depth measurements are continuously validated against geometric constraints and spatial relationships. When inconsistencies are detected (such as objects penetrating each other or impossible depth values), the system feeds back correction signals or rejects the erroneous measurements, ensuring safety-critical reliability while preserving accurate depth data where available.
2Measurement precision
If multiple frequency measurements are used to disambiguate distances, then periodic range ambiguity is resolved, but device complexity increases
Solution Approach 1:
The patent employs periodic action by using multiple frequency measurements that are systematically varied over time. Different frequencies are used in sequential measurements, allowing the system to resolve periodic range ambiguities through frequency diversity. This temporal multiplexing approach achieves high measurement precision without requiring all frequencies to be transmitted simultaneously, thus managing device complexity.
Solution Approach 2:
The patent adds a frequency dimension to the existing spatial measurement dimension. By measuring at multiple frequencies, the system creates an additional dimension of information that enables disambiguation of periodic range errors. This dimensional expansion allows precise distance measurement while the frequency dimension can be multiplexed over time, controlling overall system complexity.
3Reliability
If redundant 3D sensors are used to enhance reliability, then false positives and negatives are reduced, but device complexity and cost increase
Solution Approach 1:
The patent merges multiple 3D sensor measurements into a unified depth map by combining data from redundant sensors through fusion algorithms. This merging process leverages the complementary information from multiple sensors to reduce false positives and negatives while presenting a single consolidated output, thus achieving high detection reliability without proportionally increasing system complexity.
Solution Approach 2:
The patent designs the redundant sensor system to serve multiple functions: each sensor contributes to both primary depth detection and to validation of depth measurements. The same sensor data is used for both accurate measurement and for cross-validating reliability, making the redundant sensors multi-functional. This reduces the need for separate validation systems and manages overall device complexity.
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
This approach enhances the reliability and accuracy of 3D ToF imaging in safety-critical applications, reducing the risk of false positives and negatives, and ensuring safe operation of machinery and robots.
Implementation Method 1
Each of the at least two three-dimensional sensors is a time-of-flight (ToF) camera
Implementation Method 2
the controller is configured to disambiguate a distance to each of the illuminated objects, resolve error in received pixelwise data due to periodic range ambiguity, and determine corrected pixelwise values via each sensor illumination at the at least two different frequencies
Data Source
AI summary
An image processing system includes at least two three-dimensional sensors and a controller. Each of the sensors being for illuminating a corresponding field of view of the sensor, and generating an output array of pixelwise values indicative of distances to illuminated objects in the field of view, each sensor being configured to generate illumination at least at two different frequencies so that objects in the corresponding field of view are illuminated at the two different frequencies. The controller is communicably connected to the at least two sensors to receive pixelwise data from each sensor embodying intensity and distance information from the illuminated objects. The controller is configured to disambiguate a distance to each of the illuminated objects, resolve error in the received pixelwise data due to periodic distance ambiguity, and determine corrected pixelwise values indicative of true distance via each sensor illumination at the at least two different frequencies.