Automatic Finishing Machine Robot Teaching Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic finishing machines require skilled operators and significant time for teaching robots, leading to inefficiencies and limited applicability due to manual-based methods, which result in inaccurate and unstable finishing processes.
Innovation Solution
A method and machine that utilize three-dimensional shape data and measurement data to correct positional deviations between an ideal and actual environment, allowing for automated finishing processes without special skill or extensive teaching time, by storing shape data and teaching data in a robot's controller and using a data acquisition unit to adjust the tool's position based on measurement data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual teaching method is used to teach the robot the desired operation, then the robot can be controlled to perform finishing process, but the operator requires considerably high skill and unduly much time is required for teaching
Solution Approach 1:
The invention uses a three-dimensional model as a copy of the workpiece to generate teaching data automatically. Instead of manually teaching the robot by moving it to each position, the system creates a digital replica of the workpiece geometry and uses it to compute the optimal tool paths and positions, thereby eliminating the need for skilled operators and reducing teaching time significantly.
Solution Approach 2:
The invention replaces the manual mechanical teaching process with an automated computational system. Instead of physically moving the robot and recording positions manually, the system uses computer-based algorithms to calculate teaching data from three-dimensional model data, substituting mechanical manual operations with automated information processing.
2Extent of automation
If manual teaching method is used, then the robot can be taught to perform finishing process, but the production line must be stopped for a long period of time, leading to substantial deterioration of production efficiency
Solution Approach 1:
The invention performs preliminary generation of teaching data using three-dimensional model data before actual robot operation. By pre-computing all necessary teaching information from the digital model, the system eliminates the need for lengthy on-site teaching that would otherwise require stopping the production line, thereby maintaining high productivity while achieving automation.
3Measurement precision
If manual teaching is performed only in a range in which the robot can be controlled with eyes, then the teaching can be performed, but it is difficult to render the finishing process highly accurate
Solution Approach 1:
The system uses a three-dimensional model copy to determine precise tool positions and paths without relying on visual control during teaching. The digital model provides exact geometric information, allowing the robot to be taught positions that extend beyond the operator's visual range while maintaining high accuracy through computational precision rather than human visual estimation.
4Ease of operation
If off-line teaching system is used to prepare teaching data in advance on PC, then the need for special skill or technique for teaching operation is eliminated, but positional deviation occurs in actual environment relative to ideal environment
Solution Approach 1:
The invention incorporates feedback by measuring the actual workpiece and comparing it with the three-dimensional model to detect positional deviations. The system then uses this feedback information to correct the teaching data, compensating for differences between the ideal digital model and the actual physical environment, thereby maintaining both ease of operation and manufacturing precision.
Data Source
AI summary
This method controls an automatic finishing machine using a robot with a tool through a model storage step, a data acquisition step, a calculation step, an error derivation step, a correction step and a machining step. In the model storage step, shape data of an unfinished work or data of a three-dimensional model is stored in a memory. In the data acquisition step, the tool is brought into contact with the unfinished work W, thereby obtaining measurement data. Then, in the calculation step, actual-position data on a comparative object point is calculated based on the measurement data. Subsequently, in the error derivation step, a data difference between the calculated actual-position data and position data on the comparative object point in the three-dimensional model is obtained. Thereafter, in the correction step, teaching data indicative of the position of the tool corresponding to the shape data of the three-dimensional model is corrected, based on the data difference. Finally, in the machining step, the finishing process is performed, while the robot (11) is controlled, based on the corrected teaching data.