Camera-Based Machine Rotation State Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems require extensive sensory data and kinematics/dynamics models to determine the rotation state of machines, lacking sensor-free and accurate methods for estimating rotational motion using camera-based approaches.
Innovation Solution
A system that generates optical flow vectors from consecutively timestamped image data records, selects subsets of pixels, determines hue and intensity, and calculates a rotation angle through integral analysis of intensity plots over time, allowing for camera-based estimation of machine rotation without specific sensors or kinematics/dynamics models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional sensor-based systems are used to determine machine rotation state, then measurement accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical sensors (rotation sensors, gyroscopes, accelerometers) with a vision-based system using standard cameras. The system captures image sequences and uses optical flow analysis to estimate rotation state, substituting physical sensing mechanisms with optical field analysis. This eliminates the need for specialized sensors while maintaining measurement capability.
Solution Approach 2:
The system creates a visual representation (optical flow field) of the machine's motion state by analyzing pixel displacement patterns in image sequences. Instead of directly measuring rotation with sensors, the system copies motion information from the visual field and processes it through computational algorithms to derive rotation state, enabling sensor-free measurement.
2Device complexity
If sensor-free camera-based methods are used, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent transforms the measurement parameters from direct sensor readings to visual field characteristics. By changing from measuring physical quantities directly (angular velocity, position) to analyzing optical flow patterns (pixel displacement, hue changes, intensity variations), the system achieves accurate rotation estimation using standard cameras without specialized sensors.
Solution Approach 2:
The system transitions from one-dimensional sensor measurements to two-dimensional image plane analysis. By capturing motion information across the entire visual field and analyzing spatial-temporal patterns in image sequences, the system extracts rotation state from multiple dimensions simultaneously, improving precision despite using simpler cameras.
3Measurement precision
If extensive sensory data collection is performed, then measurement accuracy is improved, but loss of time increases
Solution Approach 1:
The system uses periodic image capture at fixed frame rates to collect motion data, replacing continuous sensor sampling. By processing discrete image sequences rather than continuous data streams, the system reduces processing overhead while maintaining sufficient temporal resolution for accurate rotation state estimation.
Solution Approach 2:
The system extracts only the essential motion information needed for rotation state estimation from image sequences, rather than collecting and processing all available sensory data. By focusing on optical flow patterns and key visual features, the system achieves accurate measurement with reduced data processing requirements and faster computation.
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 robust and accurate estimation of machine rotation states using camera data alone, reducing reliance on complex sensors and models, and correcting for estimation errors, thus improving operational precision and reducing costs in industrial settings.
Implementation Method 1
generating optical flow vectors for the pixels of the particular image data record
Data Source
AI summary
The present disclosure is directed to systems and methods for estimating machine rotation state. The rotation state can include the rotation speed and/or angle as the machine rotates. In some implementations, an estimating machine rotation state system can receive image data, timestamped consecutively, from camera devices of a machine. The estimating machine rotation state system can generate optical flow vectors for the pixels of the image data and determine the hue and intensity of the pixels based on the optical flow vectors. Using the hues, intensities, and timestamps of the image data, the estimating machine rotation state system can generate an intensity plot over time. The estimating rotation states system can then determine a rotation speed of the machine based on the intensity plot. The estimating rotation states system can also determine a rotation angle of the machine by integrating the intensity plot over time.