Transfer Robot Vision Teaching for Accurate Chamber Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for teaching transfer robots in semiconductor manufacturing are prone to errors due to operator skill level and are time-consuming, requiring manual adjustment and separate teaching marks in chambers.
Innovation Solution
A teaching method and program that utilize a vision system and substrate-type sensor to photograph mark plates and chamber structures, calculating errors and adjusting the transfer robot's position based on acquired images, eliminating the need for manual adjustment and separate teaching marks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual teaching operation is performed by operator, then teaching operation can be completed, but errors occur depending on operator skill level and time consumption increases
Solution Approach 1:
The patent replaces the manual mechanical adjustment system with an automated vision-based system. The vision chamber photographs the mark plate, and the controller automatically calculates teaching point coordinates from the captured image, eliminating the need for manual operator intervention and mechanical adjustment of the transfer robot.
Solution Approach 2:
The patent uses a mark plate with known teaching marks as a reference copy. The vision system captures an image of this mark plate, and the controller calculates the actual position by comparing the captured mark positions with their known design positions, thereby deriving accurate teaching points for the transfer robot.
2Measurement precision
If separate teaching marks are marked in the chamber, then teaching point can be derived, but device complexity and manufacturing difficulty increase
Solution Approach 1:
The patent introduces a mark plate as an intermediary object that carries teaching marks. Instead of marking the chamber directly, the mark plate is attached to a movable component (such as a chamber frame or cover) that can be positioned to bring the teaching marks into the vision chamber's field of view for automatic teaching point derivation.
Data Source
AI summary
Disclosed is a teaching method of a transfer robot including a transfer hand, the teaching method including: a first teaching operation of acquiring a first image by photographing, by a vision chamber provided on the transfer robot transferring a substrate, a mark plate attached to a chamber or a frame around the chamber, and adjusting a teaching point of the transfer robot based on the first image; and a second teaching operation of acquiring a second image by photographing, by a substrate-type sensor including an image photographing unit placed on a hand of the transfer robot, a structure in the chamber, and adjusting a teaching point of the transfer robot based on the second image.


