Robot Posture Determination Using Vision-Based Reference Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot control methods require workers to empirically determine thresholds for positional accuracy, which is time-consuming and labor-intensive, especially when precise tilt angles are necessary, and are affected by vision sensor errors.
Innovation Solution
A method involving a vision sensor to measure feature points, convert positional differences, and compare them with thresholds to determine the robot's posture, allowing for automatic correction and setting of teach points within a predetermined accuracy range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a worker determines whether the tilt of a leading end of a robot converges by calculating the circularity of a mark, then the positioning accuracy can be improved, but the workload and time required for threshold setting increases significantly
Solution Approach 1:
The patent introduces a determination point as an intermediary element between the vision sensor and the robot's leading end. Instead of directly measuring the robot's position, the system measures the position of a determination point (which is the position of the robot's leading end expressed in the vision sensor's coordinate system) and converts positional differences from the feature point to the determination point. This intermediary approach simplifies threshold setting while maintaining accuracy.
Solution Approach 2:
The patent changes the parameter being measured from the circularity of a mark (which requires complex calculation and empirical threshold setting) to the positional difference of a determination point. By expressing the robot's leading end position in the vision sensor's coordinate system and calculating positional differences directly, the system eliminates the need for complex circularity calculations and empirical threshold determination.
2Measurement precision
If a worker sets thresholds for positional difference to ensure accuracy, then the positioning precision can be improved, but the complexity of the operation increases
Solution Approach 1:
The determination point serves as an intermediary that simplifies the measurement process. Instead of requiring workers to understand and set thresholds for circularity calculations, the system uses positional difference of the determination point, which can be directly compared with simple thresholds. This makes the operation easier while maintaining precision.
Solution Approach 2:
The patent replaces the mechanical/circularity-based measurement approach with a coordinate-based approach. By expressing positions in coordinate systems and calculating positional differences mathematically, the system eliminates the need for workers to understand complex circularity calculations, making the operation simpler and more intuitive.
3Manufacturing precision
If the robot is forced to repeatedly perform motion to correct positional difference, then the positioning accuracy can be improved, but the productivity decreases due to multiple repetitions
Solution Approach 1:
The patent performs preliminary actions by pre-calculating the determination point position and establishing the coordinate transformation relationship before the repetitive correction process. This preparation work is done once, and then the system can quickly determine convergence by simply comparing positional differences, reducing the time required for each repetition and improving overall productivity.
Solution Approach 2:
The system implements feedback by continuously measuring the positional difference of the determination point and using this information to determine whether the robot has converged to the target position. This feedback mechanism allows for efficient iterative correction, reducing the number of repetitions needed while maintaining high positioning accuracy.
Data Source
AI summary
A robot system includes a robot, a vision sensor, a controller, and an input unit. The vision sensor configured to measure a feature point and obtain a measured coordinate value. The controller configured to control the robot. The input unit configured to receive an input from a user toward the controller. The controller obtains, via the input unit, setting information data on a determination point which is different from the feature point. The robot system uses a coordinate value of the determination point and the measured coordinate value, and determines whether the robot is taking a target position and orientation.