Semi-autonomous Robot Path Planning Using Vision Markers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The programming of industrial robots, such as welding robots, is often tedious and time-consuming, and prone to errors due to the need for precise calibration between simulated and actual workspaces, which can result in incorrect welds or damage to the workpiece or robot system.
Innovation Solution
A method using a vision sensor and computer vision model to detect path markers on a workpiece, determining the work path and task, and generating instructions for the robot end effector to trace the path and perform the task, allowing for semi-autonomous operation and reducing the need for extensive offline programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If offline programming is performed within a simulated environment, then the robot can be programmed without interrupting production, but calibration errors and programming mistakes can occur leading to incorrect welds or damage
Solution Approach 1:
The system uses the robot's own vision sensor to automatically detect path markers and generate programming instructions without requiring external programming equipment or manual intervention. The robot essentially programs itself by capturing images of markers and automatically translating them into motion commands, eliminating human error in programming while maintaining continuous production operation
Solution Approach 2:
The patent replaces manual mechanical programming with an automated vision-based system. Instead of requiring operators to manually input programming instructions, the system uses image capture and processing to automatically generate the robot's motion program, substituting human expertise with automated optical recognition and computational algorithms
2Ease of manufacture
If manual programming is performed, then programming can be done with existing equipment, but the process is tedious and time-consuming
Solution Approach 1:
The path markers are pre-placed on the workpiece before the robot arrives, containing all the necessary path and task information in advance. This preliminary preparation eliminates the need for time-consuming on-site programming, as the robot simply needs to capture images of the pre-placed markers to automatically generate its programming instructions
Solution Approach 2:
The system creates a digital copy of the physical workpiece markings through vision sensors. The captured images of path markers serve as data copies that contain all necessary programming information, allowing the robot to reproduce the exact path and task requirements without manual intervention or complex programming procedures
3Measurement precision
If calibration between simulated and actual workspace is performed, then programming accuracy can be maintained, but the process is complex and error-prone
Solution Approach 1:
The path markers serve as intermediary elements that bridge the gap between the physical workspace and the robot's programming system. Instead of requiring complex calibration between simulated and actual workspaces, the markers provide a simple visual reference that the robot can directly detect and interpret, eliminating the need for elaborate calibration procedures while maintaining positioning accuracy
Solution Approach 2:
The path markers use distinct visual characteristics (analogous to color changes) to convey different types of information about the work path and tasks. The vision system detects these visual differences to automatically identify marker types and extract corresponding programming instructions, simplifying the information transmission process without requiring complex calibration
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
This approach enables the robot system to semi-autonomously plan and execute complex tasks with reduced operator input, minimizing errors and saving time, while leveraging existing vision sensors and infrastructure, thus improving productivity and reducing resource consumption.
Implementation Method 1
receiving, by a device and from a vision sensor, an image of a path marker disposed on a workpiece
Data Source
AI summary
A robot controller is disclosed. The robot controller may include one or more processors, communicatively coupled to the one or more memories, configured to: receive, from a vision sensor, an image of a path marker disposed on a workpiece, the path marker having a work path identifier corresponding to a work path to be traced by a robot end effector; determine, using a computer vision model, the work path based on the image, the computer vision model having been trained to detect the work path identifier and identify the work path based on the work path identifier; determine a set of coordinates within a workspace of the robot end effector based on the work path; generate a work path instruction to guide the robot end effector according to the set of coordinates; and cause the robot end effector to operate according to the work path instruction.