3D Hand-Gesture Robot Path Teaching for Flexible Programming

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for programming industrial robots to follow complex paths and perform tasks on workpieces are cumbersome and require manual operation using teach pendants, limiting flexibility and efficiency.

Innovation Solution

A system utilizing hand gestures to generate instructions for industrial robots, where an operator uses a 3D vision sensor and computation device to interpret and convert hand gestures into robot targets and instructions, allowing the robot to perform tasks such as picking, dropping, and tool changes without pre-defined rules, enabling flexible path programming.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If manual teach pendant operation is used to program robot paths, then the robot can be programmed to follow paths, but the process is cumbersome and limits flexibility and efficiency

Engineering Contradiction:
ImproveflexibilityVSAvoidprogramming complexity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent replaces the mechanical teach pendant interface with a vision-based gesture recognition system. The 3D vision sensor captures hand gestures and converts them into robot path instructions, eliminating the need for physical teach pendant manipulation and enabling more flexible, intuitive programming.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a computation device as an intermediary between the operator's hand gestures and the robot controller. This intermediary processes gesture data, interprets intent, and generates appropriate robot instructions, enabling flexible path programming without direct mechanical interaction.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If manual teach pendant operation is used to program robot paths, then the robot can be programmed to follow paths, but the process is time-consuming and reduces efficiency

Engineering Contradiction:
Improveprogramming efficiencyVSAvoidprogramming time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The vision-based gesture recognition system replaces time-consuming manual teach pendant operations with rapid gesture capture and processing. The 3D vision sensor and computation device quickly convert hand gestures into robot paths, significantly reducing programming time and improving productivity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If hand gestures are used to program robot paths, then flexibility and programming speed are improved, but the system requires additional components such as 3D vision sensors and computation devices

Engineering Contradiction:
Improveprogramming flexibilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent integrates the 3D vision sensor and computation device into the existing robot system, allowing these components to serve multiple functions including gesture recognition, path programming, and potentially other vision-based tasks. This reduces the impact of added complexity by maximizing the utility of each component.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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 efficient and flexible programming of industrial robots to perform tasks by allowing operators to use intuitive hand gestures, reducing the need for complex manual programming and enhancing the robot's ability to adapt to different workpieces and operations.

Implementation Method 1

a 3D vision sensor to capture an image of the location pointing hand gesture

Methodology Applied
Scientific Effect3D vision sensing:

Data Source

PatentEP2931485B1Bare hand robot path teaching
Publication Date: 2023.09.13 ABB (SCHWEIZ) AG
  • EP2931485B1 patent drawingFigure 1
  • EP2931485B1 patent drawingFigure 2a~2b
  • EP2931485B1 patent drawingFigure 2c~2h

AI summary

A system for generating instructions for operating a robot to perform work on a workpiece has a 3D model or models of a robot scene that provides robot scene data. The system also has data of one or both hands of a gesture made with the one or both hands and a computation device that has program code configured to process the robot scene data and the gesture data to generate an instruction to operate the robot.