Robot End-Effector Tactile Sensing for Tool Posture Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Cooking robots face challenges in accurately identifying the type and posture of gripped tools, which affects their operational accuracy and efficiency.

Innovation Solution

A robot system equipped with an end-effector and a tactile sensor that generates three-dimensional tactile information, which is converted into a two-dimensional image for comparison with template images to determine the tool's type and posture, allowing for posture correction based on reference angles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a robot uses a tactile sensor to identify tool type and posture, then the accuracy of tool identification is improved, but the device complexity increases

Engineering Contradiction:
Improvetool identification accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces tactile sensors as intermediary devices between the robot's end-effector and the tool. These sensors act as mediators that convert physical contact information into electrical signals, enabling the robot to identify tool type and posture without requiring complex mechanical sensing mechanisms. The tactile sensor serves as an intermediary that bridges the gap between simple mechanical gripping and complex tool recognition.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the robot converts three-dimensional tactile information into two-dimensional images for processing, then the measurement precision of tool identification is improved, but the loss of information increases

Engineering Contradiction:
Improveposture detection accuracyVSAvoidtactile information loss
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent creates a two-dimensional image copy of the three-dimensional tactile information. Instead of directly processing complex 3D tactile data, the system generates a 2D representation that preserves the essential features needed for tool identification. This copying approach simplifies pattern recognition while maintaining sufficient information for accurate tool type and posture determination through template matching.

Inventive Principle:
Principle #26Copying

3Manufacturing precision

If the robot corrects the posture of the end-effector based on tool identification, then the operational accuracy is improved, but the time required for the operation increases

Engineering Contradiction:
Improveoperation accuracyVSAvoidoperation time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent implements a feedback mechanism where the robot continuously monitors tool identification through tactile sensors, compares the identified posture with the required posture, and automatically adjusts the end-effector positioning. This closed-loop feedback system enables real-time posture correction without requiring manual intervention, improving operational accuracy while minimizing time loss through automated adjustment.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11383384B2Robot and method for controlling robot
Publication Date: 2022.07.12 LG ELECTRONICS INC
  • US11383384B2 patent drawing
  • US11383384B2 patent drawing
  • US11383384B2 patent drawing

AI summary

Disclosed is a robot and a method for controlling a robot. The robot according to an embodiment of the present disclosure may include an end-effector configured to grip a tool, a tactile sensor disposed in the end-effector, the tactile sensor configured to generate tactile information about an identifier formed on the tool, and a processor configured to cause the end-effector to grip the tool, and determine at least one of a type or a posture of the tool gripped by the end-effector based on the tactile information received from the tactile sensor. Embodiments of the present disclosure may be implemented by executing an artificial intelligence algorithm and/or machine learning algorithm in a 5G environment connected for the Internet of Things.