Robot Learning Control Anti-Exception Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional learning control techniques in robots are inadequate when handling exceptional processes such as provisional stops, changes in speed override, or teaching corrections, leading to potential vibration issues, especially at high operation speeds.
Innovation Solution
Incorporating an anti-exception processing unit that reduces the robot's operation speed to a safe level and sets the learning correction amount to zero during exceptional processes, ensuring the robot's safety and preventing vibrations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If learning control is performed at high operation speeds, then productivity is improved, but vibrations occur and manufacturing precision deteriorates
Solution Approach 1:
The system performs learning control operations in advance to establish a baseline feedforward signal, then uses this pre-established signal during high-speed actual operations. The learning correction amounts calculated during preliminary learning operations are stored and applied during high-speed production, enabling high productivity while maintaining precision through the pre-optimized control parameters
Solution Approach 2:
The system periodically updates learning correction amounts by alternating between learning operations (where positional deviation is measured and correction amounts are calculated) and actual operations (where the robot performs tasks using the learned correction amounts). This periodic updating ensures that the feedforward signal remains optimized for high-speed operation while maintaining positional accuracy
2Manufacturing precision
If learning control is performed, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The control system is segmented into distinct functional units: a learning control unit that calculates learning correction amounts based on positional deviation data, a normal control unit that generates feedforward signals, and a synthesis unit that combines these signals. This segmentation allows each unit to perform its specific function efficiently, improving positional accuracy while keeping the overall system manageable through modular architecture
Solution Approach 2:
The learning correction amount acts as an intermediary element that bridges the learning control process and the normal operation. It is calculated during learning operations based on measured positional deviations and then applied as a correction to the feedforward signal during actual operations, enabling precision improvement without requiring complex real-time control modifications
Data Source
AI summary
A robot with a learning control function is disclosed. The robot includes a robot mechanism unit, a learning control unit for obtaining data on positional deviation of the robot mechanism unit upon execution of a task program and executing a learning control for calculating a learning correction amount in order to decrease the positional deviation of the robot mechanism unit below a certain value, a normal control unit for executing a learning operation of the robot mechanism unit in order to obtain the data during the learning control and executing an actual operation of the robot mechanism unit based on the learning correction amount calculated by the learning control unit after executing the learning control, and an anti-exception processing unit for executing an anti-exception process in the case where an exception process occurs during the learning operation or the actual operation.