Robot Controller Machine Learning Disturbance Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial robots face disturbances and potential failures during teaching operations due to inaccurate positioning and frictional forces, especially when operated by inexperienced users, leading to load on the robot's joints and potential failures.
Innovation Solution
A controller with a machine learning device that learns and compensates for motor disturbances by determining a compensation amount for the teaching position, using state observation, determination data acquisition, and learning sections to optimize the teaching position and reduce disturbances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If teaching operation is performed by inexperienced users, then ease of operation is improved, but manufacturing precision deteriorates due to inaccurate positioning and directional errors
Solution Approach 1:
The system performs self-diagnosis and self-correction by automatically detecting positioning errors and disturbance forces during teaching operations, then compensating for these errors through learned compensation values stored in memory, enabling the system to correct its own teaching accuracy issues without requiring expert operator intervention
Solution Approach 2:
The system implements feedback by continuously monitoring actual position data against target position data, detecting deviations and disturbance forces, storing this information in memory, and using it to generate compensation values that are applied to subsequent teaching operations, creating a closed-loop system that improves accuracy through iterative learning
2Ease of operation
If teaching operation is performed by inexperienced users, then ease of operation is improved, but reliability deteriorates due to robot failures from joint loads
Solution Approach 1:
The system prepares compensation values in advance based on detected disturbance forces and positioning errors during teaching operations, storing these compensation values in memory before subsequent operations occur, thereby cushioning against potential reliability issues by pre-correcting for errors that would otherwise cause joint loads and failures
Solution Approach 2:
The system automatically detects disturbance forces, calculates compensation values, stores them in memory, and applies them to correct positioning errors, enabling the robot to self-protect against operational failures without requiring external intervention or expert knowledge
3Manufacturing precision
If statistical processing is performed on position data to determine compensation amounts, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The system extracts only the essential compensation values needed for positioning accuracy from the collected position data, storing these extracted compensation values in memory without requiring complex statistical analysis of all raw data, thereby simplifying the control system while maintaining positioning precision
Data Source
AI summary
A machine learning device of a controller observes, as state variables expressing a current state of an environment, teaching position compensation amount data indicating a compensation amount of a teaching position in control of a robot according to the teaching position and data indicating a disturbance value of each of the motors of the robot in the control of the robot, and acquires determination data indicating an appropriateness determination result of the disturbance value of each of the motors of the robot in the control of the robot. Then, the machine learning device learns the compensation amount of the teaching position of the robot in association with the motor disturbance value data by using the observed state variables and the determination data.