Robot Position Control Using Relative Image Features for Large Moves
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot device controllers face challenges in accurately controlling the position of robots, especially when moving large distances, due to inaccuracies in calculating the Jacobian matrix and the need for prior calibration of visual sensor coordinate systems.
Innovation Solution
A robot device controller that uses a visual sensor to capture images of first and second members, processing these images to detect characteristic amounts and calculate a relative position amount, which is then used to generate movement commands for the robot to accurately position the second member relative to the first member.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a Jacobian matrix is used to calculate robot movement based on characteristic portion position differences, then the robot can be accurately driven in the vicinity of each characteristic portion, but the accuracy decreases as the robot moves away from the characteristic portion
Solution Approach 1:
The patent transitions from using a single characteristic portion (2D image coordinates) to using multiple characteristic portions distributed across the workpiece. By detecting positions of multiple characteristic portions and calculating relative position amounts between them, the system creates a multi-point reference framework that maintains accuracy across larger movement distances, effectively adding spatial dimensionality to the measurement system.
Solution Approach 2:
The patent divides the workpiece into multiple characteristic portions rather than relying on a single reference point. Each characteristic portion serves as an independent detection target, and the relative position amounts between these segmented portions are used to calculate robot movement commands, distributing the measurement function across multiple locations on the workpiece.
2Reliability
If control is performed based on a single characteristic portion in the image, then the robot can be controlled near that portion, but accuracy is low when the robot is moved by a large movement amount
Solution Approach 1:
The workpiece is segmented into multiple characteristic portions, each detectable in the image. By using multiple segmentation points rather than a single reference, the system maintains reliable position control across the entire workpiece area, enabling both large movement efficiency and high positioning reliability simultaneously.
Solution Approach 2:
The system moves from single-point control to multi-point control by detecting characteristic portions at different locations. This dimensional expansion of the reference system allows the robot to efficiently move across large areas while maintaining reliable position control through relative position calculations between multiple characteristic portions.
3Measurement precision
If a visual sensor coordinate system is calibrated with respect to a robot coordinate system, then the three-dimensional position of a workpiece can be detected, but the process requires prior calibration which increases complexity
Solution Approach 1:
Instead of performing complex coordinate system calibration, the patent uses a simplified approach by detecting two-dimensional images of characteristic portions and calculating relative position amounts directly from these image coordinates. This creates a virtual copy of the position relationship that avoids the need for physical coordinate system transformation and calibration procedures.
Solution Approach 2:
The patent extracts only the essential information needed for control - the relative position amounts between characteristic portions in the image - and discards the complex coordinate system calibration process. By taking out only the necessary relative position data rather than performing full coordinate transformation, the system achieves 3D positioning capability without the associated complexity.
Data Source
AI summary
A first characteristic portion of a first workpiece and a second characteristic portion of a second workpiece are previously determined. A characteristic amount detection unit detects a first characteristic amount related to the position of the first characteristic portion and a second characteristic amount related to the position of the second characteristic portion in an image captured by a camera. A calculation unit calculates, as a relative position amount, the difference between the first characteristic amount and the second characteristic amount. A command generation unit generates a movement command for operating a robot based on a relative position amount in the image captured by the camera and a relative position amount in a predetermined reference image.