Teach-by-Touch Robot Training for Fast Workspace Reconfiguration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Programming a general-purpose robot to perform specific tasks is a tedious process that requires communicating task goals and constraints, and reprogramming is necessary when the workspace changes, making it inefficient and skill-intensive.
Innovation Solution
A 'teach-by-touch' method where users define task goals and constraints by interacting with a projected interface in the workspace, using a device like a wand with a fiducial, allowing the robot to compute motion plans and execute tasks without manual programming, employing a robot controller with perception, planning, and interaction modules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional programming methods are used to train a robot, then the robot can execute specific tasks with precision, but the time and skill required for programming increases significantly
Solution Approach 1:
The patent replaces traditional mechanical programming approaches (manual robot handling, offline programming) with a vision-based interaction system. Users interact with the robot through natural gestures captured by cameras, and the system uses computer vision and planning algorithms to translate these gestures into executable robot commands, eliminating the need for tedious manual programming while maintaining task execution precision
Solution Approach 2:
The patent introduces an intermediary system consisting of the controller, perception module, and planning module that mediates between user gestures and robot execution. This intermediary automatically interprets user intent and generates appropriate motion plans, serving as a bridge that eliminates the need for users to directly program robot commands while ensuring accurate task execution
2Reliability
If traditional programming methods are used to train a robot, then the robot can execute specific tasks, but the complexity of the programming process increases
Solution Approach 1:
The robot system performs self-programming by automatically generating motion plans from user gestures. The perception module captures gestures, the planning module computes appropriate motions, and the robot executes them without requiring external programming intervention. This self-service capability maintains reliable task execution while eliminating programming complexity for users
Solution Approach 2:
The patent replaces complex manual programming mechanisms with an automated vision-based system. The controller uses computer vision to interpret gestures and automated planning algorithms to generate motion plans, substituting the complex mechanical process of manual programming with an intelligent automated system that maintains reliability while reducing complexity
3Adaptability or versatility
If the workspace changes, then the robot can adapt to new configurations, but reprogramming is required which reduces productivity
Solution Approach 1:
The patent makes the robot programming dynamic and adaptive to workspace changes. Instead of static pre-programming, the system allows users to define new tasks through gestures in the updated workspace configuration, and the planning module automatically adapts motion plans to the new layout. This dynamic approach maintains workspace adaptability while eliminating reprogramming time, thus preserving productivity
Solution Approach 2:
The system performs preliminary workspace analysis and motion plan generation automatically when a new task is defined. The perception module pre-processes the workspace configuration, and the planning module pre-computes appropriate motions before execution, allowing the robot to adapt to workspace changes instantly without requiring time-consuming reprogramming procedures
Data Source
AI summary
A robot-training system permits a user touch, click on or otherwise select items from a display projected in the actual workspace in order to define task goals and constraints for the robot. A planning procedure responds to task definitions and constraints, and creates a sequence of robot instructions implementing the defined tasks.