Robot Manipulator Control Gain Adjustment From Force Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Operators inexperienced in adjusting parameters such as pressing force, travel speed, and travel direction for robot manipulators face challenges, leading to inefficiencies and potential damage during operations like part fitting, gear adjustment, and polishing, as these adjustments depend heavily on manual settings rather than automated processes.
Innovation Solution
A controller system that employs machine learning, specifically reinforcement learning, to automatically adjust these parameters by detecting forces and moments applied to the manipulator, generating learning models, and optimizing control commands based on acquired data, thereby streamlining operations and reducing operator dependency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual adjustment of parameters is performed by operators, then flexibility and adaptability are maintained, but adjustment time increases and consistency deteriorates
Solution Approach 1:
The control gain adjustment system performs self-service by automatically adjusting control parameters based on force sensor data and machine learning algorithms, eliminating the need for manual operator intervention while maintaining adaptability to different operational conditions
Solution Approach 2:
The system implements continuous feedback loops where force sensor measurements are fed back to the control gain adjustment unit, which then automatically modifies control parameters in real-time based on the learned relationships between force states and optimal control gains
2Reliability
If manual adjustment of parameters is performed by operators, then experience-based optimization is possible, but operator skill dependency increases and result consistency deteriorates
Solution Approach 1:
The system replaces the mechanical dependency on operator skill and experience with an automated electronic control system that uses machine learning algorithms to determine optimal control gains, ensuring consistent high-quality adjustments regardless of operator capability
Solution Approach 2:
The control gain adjustment unit dynamically changes control parameters based on force state data and learned models, automatically optimizing parameters like pressing force, travel speed, and travel direction without requiring operator expertise
3Ease of manufacture
If manual adjustment of parameters is performed, then initial setup is possible, but risk of damage during adjustment increases
Solution Approach 1:
The system performs preliminary learning operations where the robot executes trial movements and the machine learning model accumulates force state data during these safe, controlled initial adjustments, allowing the system to learn optimal parameters before actual production work begins
Solution Approach 2:
The force sensor-based control gain adjustment provides beforehand cushioning by continuously monitoring force states and automatically adjusting control gains to prevent excessive forces that could damage workpieces or tools during the learning and operation phases
Data Source
AI summary
The controller acquires a force applied to a manipulator of a robot, to generate, based on the acquired data, force state data containing information related to the force applied to the manipulator and control command adjustment data indicating an adjustment behavior of a control command related to the manipulator as state data, thereby executing, based on the generated state data, a process of machine learning related to the adjustment behavior of the control command related to the manipulator.