Robot Abnormality Display Method for False Collision Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sensor-less collision detection methods in robots often erroneously detect collisions due to increased friction from motor or decelerator failures, leading to difficulties in identifying the actual cause of abnormalities and resulting in premature collision detection.
Innovation Solution
A method for displaying robot abnormalities that includes detecting multiple types of abnormalities, such as collision, motor lock, overload, and position deviation, with separate determination and display thresholds for each, allowing for prioritization and displaying only the most critical abnormalities when a collision is detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor-less collision detection is used to detect robot abnormalities, then collision detection accuracy is improved, but false collision detection occurs due to other abnormalities like motor lock or overload
Solution Approach 1:
The patent segments the abnormality detection process by introducing multiple independent detection functions (collision detection, motor lock detection, overload detection, position deviation detection) that operate separately but are integrated in the display phase. Each function has its own determination threshold, allowing independent evaluation of different abnormality types without interference.
Solution Approach 2:
The patent adds a new dimension to abnormality display by introducing a display determination threshold that is separate from the abnormality determination threshold. This creates a two-threshold system where abnormalities are first determined by their specific thresholds, then filtered through the display threshold to decide what to show, adding a layer of dimensional separation to the detection-display process.
2Reliability
If multiple abnormality detection functions are implemented simultaneously, then comprehensive abnormality monitoring is achieved, but it becomes difficult to identify the actual cause when collision is detected
Solution Approach 1:
The patent applies local quality by setting different display determination thresholds for different abnormality types. Each abnormality (collision, motor lock, overload, position deviation) can have its own display threshold tailored to its characteristics, allowing selective display of the most relevant abnormalities while suppressing less critical ones, thus maintaining cause identification clarity.
Solution Approach 2:
The patent changes parameters by introducing display determination thresholds that can be adjusted independently from the abnormality determination thresholds. This allows dynamic control over which abnormalities are displayed based on their severity and relevance, transforming the fixed threshold system into a flexible parameter-based display control mechanism.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In sensor-less collision detection of a robot, a conventional method for displaying abnormality of the robot makes an abnormality display of "collision detected" when a collision is erroneously detected, and does not clarify the situations under which the collision has been erroneously detected. In a method for displaying abnormality of a robot, when a collision of the robot is detected and the collision detection is displayed, at least one abnormality display is selected from a plurality of abnormality detection items different from collision detection. This can offer information useful for the user to understand the situations at occurrence of the erroneous collision detection.