Robot Teaching Correction Using Dual Feature Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot teaching systems face inefficiencies due to individual differences in workpiece features and feature obfuscation, leading to reduced detection rates and the need for manual correction, which increases the time required for teaching operations.
Innovation Solution
A robot system that includes an imaging device, a robot controller, and a storage unit for pre-stored features and positional relationships, utilizing feature detection and calculation sections to automatically correct the robot's teaching position and posture based on detected features, even when primary features are obscured, by using secondary features for detection when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If feature detection is performed on workpieces with individual differences or obfuscated features, then detection reliability decreases, but manual correction increases time consumption
Solution Approach 1:
The patent segments the feature detection process into multiple independent detection targets (first feature and second feature). The system detects the first feature primarily, but when detection fails or reliability is insufficient, it automatically falls back to detecting the second feature. This segmentation approach ensures continuous reliable operation without manual intervention, resolving the contradiction between detection reliability and time loss.
Solution Approach 2:
The patent prepares backup detection features (second feature) in advance alongside the primary feature (first feature). By pre-configuring alternative detection targets with known positional relationships, the system cushions against detection failures caused by individual workpiece differences or obfuscation. This beforehand preparation eliminates the need for manual correction while maintaining high detection reliability.
2Adaptability or versatility
If multiple feature templates are stored in advance for different workpiece types, then detection coverage improves, but teaching operation becomes more time-consuming
Solution Approach 1:
The patent creates a universal detection system where the robot controller can detect multiple types of features (first feature and second feature) using the same detection mechanism. The controller stores positional relationships between these features and the target position, allowing it to adapt to different workpiece types without requiring separate teaching operations for each type. This multi-functionality approach improves detection coverage while avoiding the time consumption associated with teaching multiple specialized systems.
Solution Approach 2:
The patent performs preliminary action by pre-storing the positional relationships between features and target positions in the robot controller. This preliminary preparation allows the system to automatically calculate correct target positions during actual operation without requiring time-consuming teaching operations for each workpiece type. The controller uses these pre-stored relationships to rapidly adapt to different workpiece configurations.
3Measurement precision
If manual correction of robot teaching is performed for undetected features, then teaching accuracy is maintained, but productivity decreases
Solution Approach 1:
The patent implements self-service by enabling the robot system to automatically correct its own teaching positions using image processing and feature detection. When the first feature cannot be detected or reliability is insufficient, the system automatically switches to detecting the second feature and recalculating the target position without requiring manual intervention. This self-service mechanism maintains teaching accuracy while significantly improving productivity by eliminating manual correction steps.
Solution Approach 2:
The patent incorporates feedback mechanisms where the robot controller continuously monitors feature detection results and automatically adjusts teaching positions based on detected feature positions. The system uses image processing to obtain actual workpiece positions and feeds this information back to correct the teaching data. This feedback loop ensures high teaching accuracy while maintaining high productivity through automated correction rather than manual intervention.
Data Source
AI summary
A robot system includes a target position calculation section which calculates, when a first feature can be detected from an image, a target position of a robot based on the calculated position of the first feature and a stored first positional relationship, and calculates, when the first feature cannot be detected from the image and a second feature can be detected from the image, a target position of the robot based on the calculated position of the second feature and the stored first positional relationship.