Robot Hand Grasp Detection Using Friction Coefficient Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing grasp detection techniques for robot hand devices face challenges in accurately detecting thin-film objects, especially when they are in a grasped state, due to difficulties with vision sensors in low light conditions, tactile sensors struggling with small objects hidden by fingertips, and the inability to differentiate between grasped and non-grasped states, particularly when objects are thin-film-like or too small.
Innovation Solution
A control device that includes force detecting means for normal and shearing forces, and object detecting means to calculate friction coefficients, allowing for the detection of thin-film objects by comparing these coefficients with a threshold value, and optionally using position detecting means to differentiate between thin-film and non-thin-film objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If vision sensors are used for grasp detection, then detection capability is improved, but detection accuracy deteriorates in low light conditions
Solution Approach 1:
The patent replaces vision-based detection with a tactile sensor system that measures physical forces. The tactile sensor detects normal force and shearing force components, substituting optical detection with mechanical force measurement to achieve environment-independent grasp detection.
Solution Approach 2:
The patent introduces friction coefficient calculation as an intermediary parameter between force measurement and grasp detection. By computing the friction coefficient from normal and shearing forces, the system creates a reliable intermediate metric that accurately indicates grasp state regardless of lighting conditions.
2Difficulty of detecting and measuring
If tactile sensors are used for grasp detection, then detection capability is improved, but detection accuracy deteriorates for small thin-film objects hidden by fingertips
Solution Approach 1:
The patent changes the detection parameter from simple contact presence to friction coefficient calculation. By deriving the friction coefficient from the ratio of shearing force to normal force, the system can detect thin-film objects that are hidden by fingertips, as the friction coefficient changes characteristically when such objects are grasped.
Solution Approach 2:
The patent performs preliminary calibration to establish the friction coefficient characteristics of thin-film objects before actual detection. The system pre-stores the friction coefficient range of thin-film objects and uses this reference information to accurately identify when such objects are grasped, even when hidden from direct view.
3Device complexity
If simple contact detection is used, then device complexity is reduced, but detection accuracy deteriorates in distinguishing grasped vs non-grasped states
Solution Approach 1:
The patent performs preliminary calibration to establish the friction coefficient characteristics of thin-film objects before actual detection. The system pre-stores the friction coefficient range of thin-film objects and uses this reference information to accurately identify when such objects are grasped, even when hidden from direct view.
Solution Approach 2:
The patent changes the detection parameter from simple contact presence to friction coefficient calculation. By deriving the friction coefficient from the ratio of shearing force to normal force, the system can detect thin-film objects that are hidden by fingertips, as the friction coefficient changes characteristically when such objects are grasped.
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 accurate detection of thin-film objects in a grasped state by distinguishing between maximum stationary friction coefficients and those of the thin-film objects, improving the robot hand device's ability to handle and manipulate thin-film objects effectively.
Implementation Method 1
force detecting means for detecting a force in a normal direction and a force in a shearing direction of fingertips of a robot hand device respectively as a normal force and a shearing force, and object detecting means for calculating a friction coefficient using the normal force and the shearing force
Data Source
AI summary
A control device includes a force detector configured to detect a force in a normal direction and a force in a shearing direction of fingertips of a robot hand device respectively as a normal force and a shearing force, and an object detector configured to calculate a friction coefficient using the normal force and the shearing force detected by the force detector and to detect whether or not a thin-film object having a maximum friction coefficient different from a maximum stationary friction coefficient between the fingertips is grasped between the fingertips on the basis of the calculation result.


